Intelligent agent control method and device, equipment and storage medium
By dynamically adjusting the number and combat power of agents in the simulation game and controlling the behavior of agents according to the real-time combat loss rate, the problems of existing agent behavior fixity and behavior tree complexity are solved, and higher flexibility and Resource optimization effect.
Patent Information
- Application Number
- CN202411418107.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-11
- Publication Date
- 2025-02-21
AI Technical Summary
Existing agents trained through behavioral trees are difficult to dynamically modify and adapt to new situations and changing game states during runtime, resulting in increased exposure risk of behavioral fixity, and the complex structure of the behavioral tree is difficult to maintain and expand.
By obtaining game character information in the simulation game, the number and combat power of agents are generated, and the behavior and quantity of agents are adjusted according to the real-time combat loss rate, and dynamic surrender is achieved to control combat loss rate.
It improves the flexibility and dynamic performance of the agent, reduces the complexity and maintenance difficulty of the behavior tree, and enhances the adaptability and resource utilization of the system.
Smart Images
Figure CN119488720A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to an intelligent body control method, device, equipment and storage medium. Background Art
[0002] In recent years, with the continuous development of the game industry, the design of game agents based on artificial intelligence (AI) has become an indispensable part of game development. As a game category that emphasizes development over competition, the goal of the agents in simulation games (SLG) is to provide players with a better gaming experience. In particular, in competitive scenarios such as SLG stronghold battles, the goal of the agent is not to defeat human players, but to ensure that human players suffer a certain degree of loss while allowing the players to win, so as to enhance the players' gaming experience and increase player stickiness.
[0003] The existing intelligent agents trained by behavior trees have fixed behaviors and decision logic defined during design, which makes it difficult to dynamically modify and adapt to new situations and changing game states at runtime. The fixed nature of the intelligent agent's behavior also increases the risk of exposure to the intelligent agent. With the expansion of the behavior tree and the need to add new nodes, its structure may become complex, chaotic, and coupled, which may make the behavior tree difficult to maintain in the future and difficult to further expand.
[0004] Therefore, there is an urgent need for an intelligent agent control method that can greatly improve the flexibility and dynamic performance of the intelligent agent, so that the intelligent agent can dynamically adapt to the player's behavior and take corresponding actions in an SLG environment. Summary of the invention
[0005] The present application provides an agent control method, device, equipment and storage medium, which aims to solve the problem that the existing agents trained by behavior trees have fixed behaviors and decision logic defined during design, which makes it difficult to dynamically modify and adapt to new situations and changing game states during runtime, and the fixed nature of the agent's behavior increases the risk of exposure of the agent. With the expansion of the behavior tree and the need to add new nodes, its structure may become complex, chaotic, and coupled, which may lead to the problem that the behavior tree is difficult to maintain in the later stage and difficult to further expand.
[0006] In a first aspect, the present application provides an intelligent agent control method, comprising:
[0007] Obtaining game character information to be played in the simulated game, where the game character information at least includes the number of game characters and the combat power of the game characters;
[0008] Generate the number of intelligent entities and their combat power according to the number of game characters and their combat power;
[0009] Generate multiple agents according to the number of agents and the combat power of the agents, and add the generated agents to the simulation game to start the game of the simulation game;
[0010] Generate a target damage rate based on the number of agents, agent combat power, number of game characters, and game character combat power;
[0011] Get the current combat loss rate of the game characters in the simulated game;
[0012] If the current loss rate is greater than the target loss rate, the agent is controlled to surrender in the simulated game.
[0013] In some embodiments, the target loss rate includes a stage loss rate and a total loss rate, and the total loss rate is greater than the stage loss rate; if the current loss rate is greater than the target loss rate, the intelligent agent is controlled to surrender in the game of the simulation game, including: if the current loss rate is less than the stage loss rate, updating the number of game characters, the combat power of the game characters, the number of intelligent agents, and the combat power of the intelligent agents; adjusting the number of intelligent agents and the combat power of the intelligent agents according to the updated number of game characters, the combat power of the game characters, the number of intelligent agents, and the combat power of the intelligent agents, and restarting the game of the simulation game; updating the current loss rate until the current loss rate is greater than the stage loss rate; if the current loss rate is greater than the total loss rate, controlling the intelligent agent to surrender in the game of the simulation game.
[0014] Exemplarily, adjusting the number of agents and the combat power of agents according to the updated number of game characters, the combat power of game characters, the number of agents and the combat power of agents includes: generating a combat power value of game characters according to the number of game characters and the combat power of game characters; generating an agent combat power value according to the number of agents and the combat power of agents; obtaining the current number of rounds and the expected number of rounds of the simulated game; adjusting the number of agents and the combat power of agents according to the current number of rounds and the expected number of rounds of the game according to the combat power value of the game characters and the combat power value of the agents, and the expressions corresponding to the adjustment of the number of agents and the combat power of agents include:
[0015]
[0016] Among them, F aI,control Indicates adjusting the combat power value of the intelligent body, N AI,control represents the number of agents after adjustment, P AI，cobtrol Indicates the adjusted combat power of the intelligent agent, F AI,max Indicates the combat power of the intelligent agent, F u,max Indicates the game character's combat power value, n1 indicates the current round number, and n2 indicates the expected round number.
[0017] Exemplarily, if the current loss rate is less than or equal to the total loss rate, the method further includes: updating the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters; updating the stage loss rate according to the current loss rate, the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters, and restarting the simulation game until the current loss rate is greater than the stage loss rate and the total loss rate.
[0018] In some embodiments, a target loss rate is generated based on the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters, including: obtaining the expected number of rounds of the simulated game; generating a corresponding single loss ratio of the simulated game based on the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters; and obtaining the target loss rate based on the expected number of rounds and the single loss ratio.
[0019] Exemplarily, obtaining a target loss rate according to an expected number of game rounds and a single loss ratio includes: calculating the target loss rate according to a preset target loss rate calculation formula, the expected number of game rounds, and a single loss ratio; wherein the target loss rate calculation formula includes:
[0020]
[0021] Among them, n2 represents the expected number of rounds, Indicates the single battle damage ratio.
[0022] In some embodiments, the number of intelligent agents and their combat power are generated according to the number of game characters and their combat power, including: obtaining difficulty information corresponding to the simulation game; obtaining the number generation coefficient and the combat power generation coefficient corresponding to the intelligent agent according to the difficulty information; generating the number of intelligent agents according to the number of game characters and the number generation coefficient; generating the combat power of the intelligent agents according to the combat power generation coefficient and the combat power of the game characters.
[0023] In a second aspect, the present application also provides an intelligent body control device, comprising:
[0024] An information acquisition module is used to acquire information about game characters to be played in the simulated game, where the game character information at least includes the number of game characters and the combat power of the game characters;
[0025] A combat power generation module is used to generate the number of intelligent entities and their combat power according to the number of game characters and their combat power;
[0026] An agent generation module is used to generate multiple agents according to the number of agents and the combat power of the agents, and add the generated agents to the simulation game to start the game of the simulation game;
[0027] A target generation module is used to generate a target combat loss rate based on the number of agents, the combat power of the agents, the number of game characters, and the combat power of the game characters;
[0028] The battle damage acquisition module is used to obtain the current battle damage rate corresponding to the game character in the game of the simulation game;
[0029] The control completion module is used to control the agent to surrender in the simulation game if the current loss rate is greater than the target loss rate.
[0030] In a third aspect, the present application further provides a computer device, including:
[0031] Memory and processor;
[0032] The memory is used to store computer programs;
[0033] The processor is used to execute the computer program and implement the steps of the intelligent agent control method as described in the first aspect above when executing the computer program.
[0034] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the steps of the intelligent agent control method as described in the first aspect above.
[0035] The provided method obtains and analyzes the information of the game characters to be played before the start of the simulation game. This information includes: the number of game characters. The combat power of the game characters (their strength level, ability, etc.). Based on the obtained number and combat power information of the game characters, the system generates a corresponding number of agents and assigns appropriate combat power to each agent. This step ensures that the generation of the agent is consistent with the current state of the game environment and supports the rationality of subsequent simulations. The initial configuration of the agent is completed by adding the generated multiple agents to the simulation game scene. After the preparation phase is over, the game of the simulation game can be launched. Using the number and combat power of the agents and game characters, the target combat loss rate for this game is calculated and set. The target combat loss rate is a standard for evaluating the game results and the performance of the agents. In the simulation game, the current combat loss rate of the game characters is obtained in real time. The actual combat loss rate is compared with the pre-set target combat loss rate. If the current combat loss rate exceeds the target combat loss rate, the agent will take a "surrender" behavior to terminate the game. This step ensures that the battle results are within a controllable range to reduce unnecessary waste of resources.
[0036] The proposed method has at least the following beneficial effects:
[0037] 1. Improved flexibility: The method dynamically adjusts the behavior and number of agents according to the real-time situation, overcoming the static limitations of the behavior tree and improving the flexibility and adaptability of the system.
[0038] 2. Complexity management: Adopting decision logic guided by combat damage rate avoids complex behavior tree structure, making subsequent maintenance and expansion easier.
[0039] 3. Risk control: By monitoring the loss rate, the agent can automatically terminate the game when the risk is too high, reducing the exposure risk caused by bad strategies.
[0040] 4. Resource optimization: The intelligent agent makes behavioral decisions based on reasonable calculation and configuration, which improves resource utilization and the overall efficiency of the system.
[0041] In summary, the provided method can form a variety of battle rhythms according to the settings. While meeting the flexibility and dynamics of the AI-trained agent, it is lighter and easier to maintain than the behavior tree as a whole. At the same time, it can make the AI agent's strategy more humane in scenes such as SLG stronghold competition, reduce the risk of AI exposure, and dynamically adapt to the combat power of human players to ensure that the enthusiasm of human players is not dampened while improving the gaming experience of human players. It can also dynamically and adaptively adapt the business's adjustment of the battle loss ratio without retesting the parameters of the adaptation settings. The provided method breaks through the inherent limitations of the behavior tree and manages the agent behavior in a more dynamic and intelligent way, thereby improving the operating efficiency of the simulation game and the performance of the agent.
[0042] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0044] Figure 1 is a schematic flow chart of the steps of an intelligent agent control method provided by an embodiment of the present application;
[0045] Figure 2 This is a schematic diagram of the principle of an SLG game provided by an embodiment of the present application;
[0046] Figure 3 is a flow chart of an intelligent agent control method provided by an embodiment of the present application;
[0047] Figure 4 is a schematic diagram of the structure of the intelligent body control device provided in an embodiment of the present application;
[0048] Figure 5It is a schematic block diagram of the structure of the computer device provided in the embodiment of the present application.
[0049] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION
[0050] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0051] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.
[0052] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.
[0053] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0054] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0055] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.
[0056] In recent years, with the continuous development of the game industry, the design of game agents based on artificial intelligence (AI) has become an indispensable part of game development. As a game category that emphasizes development over competition, the goal of the agents in simulation games (SLG) is to provide players with a better gaming experience. In particular, in competitive scenarios such as SLG stronghold battles, the goal of the agent is not to defeat human players, but to ensure that human players suffer a certain degree of loss while allowing the players to win, so as to enhance the players' gaming experience and increase player stickiness.
[0057] The existing intelligent agents trained by behavior trees have fixed behaviors and decision logic defined during design, which makes it difficult to dynamically modify and adapt to new situations and changing game states at runtime. The fixed nature of the intelligent agent's behavior also increases the risk of exposure to the intelligent agent. With the expansion of the behavior tree and the need to add new nodes, its structure may become complex, chaotic, and coupled, which may make the behavior tree difficult to maintain in the future and difficult to further expand.
[0058] Therefore, there is an urgent need for an intelligent agent control method that can greatly improve the flexibility and dynamic performance of the intelligent agent, so that the intelligent agent can dynamically adapt to the player's behavior and take corresponding actions in an SLG environment.
[0059] To solve the above problems, please refer to Figure 1 , Figure 1 : is a schematic flow chart of an intelligent agent control method provided by an embodiment of the present application. The intelligent agent control method can be implemented by a computer device, and the computer device can be deployed on a single server or a server cluster. It can also be deployed on a handheld terminal, a laptop computer, a wearable device or a robot, etc.
[0060] At the same time, behavior trees define fixed behaviors and decision logic during design, which makes it difficult to dynamically modify and adapt to new situations and changing game states at runtime, resulting in limited flexibility of the agent, while the fixed nature of the agent's behavior increases the risk of exposure to the agent. As the behavior tree expands and new nodes are added, its structure may become complex, chaotic, and coupled, which may make the behavior tree difficult to maintain later and difficult to expand further.
[0061] At the same time, if Figure 2As shown, the method provided by the present application is applied in a simulation game (SLG), where multiple human players gather to create a union, and a stronghold war refers to a war between unions (GVG), the goal of which is to compete for and control a specific stronghold or territory in the game world. Stronghold wars usually have the following characteristics: 1. Stronghold competition: Union members compete for specific strongholds through attack and defense. These strongholds may be cities, fortresses, fortresses or other important geographical locations. 2. Resources and rewards: Strongholds are usually associated with specific resources, rewards and privileges. Competing for and controlling strongholds can provide in-game resources, experience points, equipment or other advantages for unions and union members. 3. Collaboration and strategy: A successful stronghold war requires effective collaboration and strategy among union members. Union members may need to develop attack plans, dispatch troops, coordinate defenses, and conduct diplomacy and negotiations with other unions. 4. Offensive and defensive battles: Stronghold wars involve attack and defense. When the union occupying the stronghold is not our own union, union members need to prepare to attack the strongholds occupied by the enemy; when the union occupying the stronghold is our own union, at the same time, defenses need to be arranged to protect their own occupied strongholds. Depending on the outcome of each battle, both the winner and the loser will suffer a fixed percentage of casualties.
[0062] like Figure 1 As shown, the intelligent agent control method provided in this embodiment includes steps S101 to S106. The details are as follows:
[0063] Step S101. Obtain information about game characters to be played in the simulated game, where the game character information at least includes the number of game characters and the combat power of the game characters.
[0064] Specifically, in general, before a stronghold war such as in SLG is about to start, union members will mention going to the stronghold to prepare for the war. Therefore, before the war starts, based on the participating members of each union near the current stronghold, agents with matching numbers and combat power can be assigned to go to the stronghold to avoid too many agents going to attack or too strong combat power, which will undermine the enthusiasm of humans to attack. The method extracts game character information from the corresponding SLG database or real-time monitoring tool by setting a data acquisition module. Including the number and combat power of each character, and even other features such as character type, skills, etc. For example, a set of API interfaces is used to regularly obtain character data from the game server, or the character information is automatically extracted before each game starts through the built-in script of the game. To provide accurate and necessary data support for the subsequent agent generation, while ensuring the matching degree between the agent strategy and the current game status.
[0065] Step S102. Generate the number of intelligent agents and the combat power of the intelligent agents according to the number of game characters and the combat power of the game characters.
[0066] Specifically, by setting the algorithm, the number of agents may be adjusted based on a ratio (e.g., one agent is generated for every two characters). The character power is converted into the appropriate agent power by using a linear or nonlinear mapping method. For example, an algorithm module is developed to input the number of characters and power, and output the number of agents and the corresponding power fitting value. Then, the agent attributes are adjusted based on real-time data to ensure that they perform reasonably in the game. A balance of capabilities is maintained between characters and agents to improve the player's gaming experience and reduce the risk of agent exposure.
[0067] In some embodiments, the number of intelligent agents and their combat power are generated according to the number of game characters and their combat power, including: obtaining difficulty information corresponding to the simulation game; obtaining the number generation coefficient and the combat power generation coefficient corresponding to the intelligent agent according to the difficulty information; generating the number of intelligent agents according to the number of game characters and the number generation coefficient; generating the combat power of the intelligent agents according to the combat power generation coefficient and the combat power of the game characters.
[0068] In this embodiment, the provided method first needs to determine the difficulty level of the current simulation game. The difficulty information may be set by the initial design of the game, or it may be dynamically adjusted according to the current game process. The difficulty may be expressed as multiple levels such as simple, medium, and difficult, and each level corresponds to a different coefficient. Based on the difficulty information obtained, the method extracts the number generation coefficient and the combat power generation coefficient of the agent from the predefined configuration. These coefficients determine how to adjust the number and combat power of the agent under different difficulties. For example, a simple difficulty may have a lower coefficient, while a high difficulty corresponds to a higher coefficient, thereby increasing the challenge. Using the number of game characters and the number generation coefficient, the number of agents that need to be generated is calculated. If the number of agents is set = the number of game characters × the number generation coefficient, it can ensure that the number of agents is proportional to the number of game characters.
[0069] At the same time, the combat power of the intelligent body is calculated and generated according to the combat power generation coefficient and the combat power of the game character. The formula is, for example: intelligent body combat power = game character combat power × combat power generation coefficient. Thus, this solution can achieve a good balance between ensuring challenge and playability.
[0070] By dynamically adjusting the number and combat power of agents according to the difficulty of the game, the game difficulty can be automatically adjusted by the system to match the player's operating level or design intent. While ensuring the ability balance between game characters and agents, different challenging experiences are provided according to the difficulty level. By accurately adjusting the agent parameters, the challenge of the game is maintained while avoiding the frustration of players caused by excessive difficulty. Designers can quickly match different game experiences by adjusting these coefficients, which is convenient for testing and optimizing game balance.
[0071] In general, this embodiment improves the efficiency and rationality of agent generation in simulation games through a systematic adjustment method, allowing the game to be adaptively adjusted according to various factors, thereby improving the overall gaming experience.
[0072] Step S103. Generate multiple intelligent agents according to the number of intelligent agents and their combat power, and add the generated intelligent agents to the simulation game to start the game of the simulation game.
[0073] Specifically, the agent object is automatically constructed by calling the agent generation engine or module. For example, a Python script is used in combination with the game engine API to automatically generate the set agent and load it into the game in the form of a configuration file. By automatically setting the initial position, state, target and other attributes of the agent, the scene can be quickly set at the beginning of the game, saving time. At the same time, the unified generation standard improves the fairness of the game.
[0074] Step S104. Generate a target combat loss rate based on the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters.
[0075] Specifically, by designing statistical models or using machine learning algorithms, appropriate target damage rates are derived based on the input agent and role information. For example, historical game data is used to train a regression model, current game parameters are input, and the predicted target damage rate is output. This standardizes the evaluation of game results and facilitates subsequent analysis. At the same time, the target damage standard is continuously optimized based on historical data and actual experience.
[0076] Step S105: Obtain the current combat loss rate corresponding to the game character in the game of the simulation game.
[0077] Specifically, a real-time analysis module is built to continuously obtain and calculate the current damage rate. The damage data in the game, such as the number of deaths and damage values, are integrated into an easily calculable indicator. For example, through the real-time data stream interface, the character damage and loss data are obtained through the event monitoring mechanism and the real-time damage rate is calculated. This ensures that key information is quickly obtained in the changing game state. This makes the agent's decision accurate and well-founded.
[0078] For example, since it is impossible to know the total number of soldiers of the enemy union corresponding to the agent in SLG, the total combat loss rate of the enemy union is defined as the current total combat loss and the current total number of exposed soldiers. The specific calculation method is shown in the following formula:
[0079] Step S106. If the current loss rate is greater than the target loss rate, the agent is controlled to surrender in the simulation game.
[0080] Specifically, by setting a conditional trigger mechanism, the agent triggers surrender behavior when the loss rate exceeds the standard. Surrender is achieved by setting AI module instructions for the agent, which is manifested as weakening combat power, stopping attacking or exiting the battle. For example, the surrender condition is introduced in the agent logic tree using Boolean logic conditions, and the surrender function is automatically called when the loss rate exceeds the threshold. While ensuring that the enthusiasm of human players is not discouraged, the gaming experience of human players is improved, and unnecessary games are ended in time to save system resources. Avoid meaningless consumption when you are clearly at a disadvantage. Through the implementation of the above steps, the provided method can significantly improve operating efficiency and adaptability, and can be applied in various dynamically changing game scenarios to demonstrate excellent capabilities and strategy adjustments.
[0081] In some embodiments, the target loss rate includes a stage loss rate and a total loss rate, and the total loss rate is greater than the stage loss rate; if the current loss rate is greater than the target loss rate, the intelligent agent is controlled to surrender in the game of the simulation game, including: if the current loss rate is less than the stage loss rate, updating the number of game characters, the combat power of the game characters, the number of intelligent agents, and the combat power of the intelligent agents; adjusting the number of intelligent agents and the combat power of the intelligent agents according to the updated number of game characters, the combat power of the game characters, the number of intelligent agents, and the combat power of the intelligent agents, and restarting the game of the simulation game; updating the current loss rate until the current loss rate is greater than the stage loss rate; if the current loss rate is greater than the total loss rate, controlling the intelligent agent to surrender in the game of the simulation game.
[0082] In the attack strategy of the intelligent agent in the competition phase of SLG, the first thing to do is to set the number of stages for dividing the entire battle, and set a target loss rate for each stage. If the current total loss rate reaches the target loss rate, the battle enters the next stage. At the same time, a final loss target needs to be set as a trigger condition for giving up the battle and letting humans win. When the human union reaches the final loss target, the intelligent agent chooses to give up the battle and retreat.
[0083] like Figure 3 As shown. In the simulation game, the target loss rate is divided into a stage target (stage loss rate) and a total target (total loss rate). The total loss rate is always greater than the stage loss rate. If the current loss rate exceeds the stage loss rate but is lower than the total loss rate, the provided method will update the game parameters (such as the number and combat power of characters and agents) and restart the game to make adjustments. If the current loss rate exceeds the total loss rate, the provided method will immediately control the agent to surrender to end the game.
[0084] By introducing the phased casualty rate, the method can make multiple adjustments before reaching the total casualty rate, and gradually optimize the game settings. At the same time, it reduces unnecessary consumption, avoids the possibility of premature termination, and improves resource utilization through timely adjustments. At the same time, the application of phased strategies makes the decision-making process more flexible, and can make gradual adjustments to multiple rounds of games.
[0085] For example, the stage loss rate is usually used to control the combat performance in the short term, so it is necessary to set the game rhythm and corresponding goals for each stage. This setting may be based on the game scene, level arrangement or specific combat mission. Using the historical data of the game, analyze the performance of the loss rate in different periods, and understand the reasonable range of losses in each stage through statistical analysis. The target loss rate to be achieved can be derived from this analysis, making the stage goal more practical and operational. As needed, adjustment factors can be introduced, such as the difficulty of the performance of recent game players, to dynamically adjust the stage loss rate. These factors can help adapt to the changing player level and challenge requirements. The expression of the stage loss rate may be a function f()=f(game character combat power, intelligent agent combat power, level difficulty), which may include the combat power comparison of specific characters and intelligent agents in this stage, plus adjustment factors. The total loss rate represents the acceptable overall loss level in the entire game process and should match the overall game goal. This usually includes expectations for the game ending, such as pass or fail conditions. The overall total loss rate is accumulated based on the loss rates of each stage to generate an overall total loss rate. This process requires comprehensive consideration of the contribution of each stage to the overall situation. For example, some key stages may require additional weights. Based on real-time game data and player feedback, the level of total immediate damage is gradually observed, and adjustments are made between relaxed or strict standards. The entire process achieves a more objective assessment by observing the accumulated loss data. The total damage rate may be obtained in a manner similar to probability accumulation or loss measurement (such as the sum of the product of the damage rate of each stage and its weight), ensuring that the impact of each stage on the total loss is comprehensively considered.
[0086] Through these steps, the phase and total loss rates can be generated and adjusted in a dynamic environment to ensure a reasonable challenge level and experience balance in each game phase and throughout the game. Through continuous monitoring and adjustment, these loss rates can also be further optimized to adapt to actual conditions.
[0087] Exemplarily, adjusting the number of agents and the combat power of agents according to the updated number of game characters, the combat power of game characters, the number of agents and the combat power of agents includes: generating a combat power value of game characters according to the number of game characters and the combat power of game characters; generating an agent combat power value according to the number of agents and the combat power of agents; obtaining the current number of rounds and the expected number of rounds of the simulated game; adjusting the number of agents and the combat power of agents according to the current number of rounds and the expected number of rounds of the game according to the combat power value of the game characters and the combat power value of the agents, and the expressions corresponding to the adjustment of the number of agents and the combat power of agents include:
[0088]
[0089] Among them, F AI,control Indicates adjusting the combat power of the intelligent body, N AI,control represents the number of agents after adjustment, PAI，control Indicates the adjusted combat power of the intelligent agent, F AI,max Indicates the combat power of the intelligent agent, F u,max Indicates the game character's combat power value, n1 indicates the current round number, and n2 indicates the expected round number.
[0090] Furthermore, this patent proposes a method for adapting the combat power of an expedition based on a reduction coefficient and the expected number of hits. This method will gradually increase its own combat power in a round of battle to avoid the lack of gaming experience for human players while avoiding the over-strike enthusiasm in the first attack. The combat power value of the game character and the combat power value of the intelligent agent are calculated according to the updated parameters. Obtain the information on the number of rounds of the game: obtain the current number of rounds of the game and the expected number of rounds of the game to guide the adjustment of the strategy. Adjustments will be made based on the relative combat power of the character and the intelligent agent and the expected number of rounds of the game. Through mathematical modeling of influencing factors, precise control of the number and combat power of intelligent agents can be achieved. At the same time, dynamic adjustments are made based on the combat power and the number of rounds of the game to ensure that the intelligent agent is always in good condition to cope with game changes. It can also reduce invalid attempts caused by inappropriate game parameters and improve the effectiveness of strategy implementation.
[0091] Exemplarily, if the current loss rate is less than or equal to the total loss rate, the method further includes: updating the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters; updating the stage loss rate according to the current loss rate, the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters, and restarting the simulation game until the current loss rate is greater than the stage loss rate and the total loss rate.
[0092] When the current loss rate is lower than or equal to the total loss rate, continue to update the number and combat power of agents and characters to ensure continuous optimization of the game. According to the real-time loss rate and updated game parameters, adjust the stage loss rate to make it more suitable for the current game status. Continue to monitor and update the game status until the current loss rate exceeds the updated stage loss rate and total loss rate. Even if the loss does not exceed the standard, continue to optimize the game strategy and enhance the system's ability to cope with changes. By continuously adjusting the stage loss rate, improve the adaptability and flexibility of the system in the game. Through dynamic adjustment, ensure that the game experience is not completely unbalanced due to a single mistake.
[0093] The above embodiments and their examples constitute a complete intelligent agent dynamic management system. Through sophisticated strategy adjustment and real-time monitoring, the performance of the intelligent agent and the gaming experience are significantly improved.
[0094] In some embodiments, a target loss rate is generated based on the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters, including: obtaining the expected number of rounds of the simulated game; generating a corresponding single loss ratio of the simulated game based on the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters; and obtaining the target loss rate based on the expected number of rounds and the single loss ratio.
[0095] First, the method provided needs to clearly define the expected number of rounds in the simulated game, which may be determined based on player experience, historical data, or game settings. Based on the number and combat power information of agents and game characters, the system calculates the acceptable single-shot damage ratio for each round of the game. This ratio reflects the degree of loss of the agent in a single round of confrontation. Combining the expected number of rounds and the single-shot damage ratio, the system obtains the overall target damage rate through cumulative calculation. This target value represents the overall damage level that the agent can tolerate within all expected rounds.
[0096] Then, by calculating the expected number of rounds and the single-round tolerance, the system can set a clear target for the casualty rate and improve the clarity of the strategy. At the same time, it provides a measurable indicator for the agent's behavioral decision-making to guide adjustments and optimizations during the game. The single-round casualty ratio is dynamically adjusted as the game progresses, so that the overall strategy can remain effective in different contexts.
[0097] Exemplarily, obtaining a target loss rate according to an expected number of game rounds and a single loss ratio includes: calculating the target loss rate according to a preset target loss rate calculation formula, the expected number of game rounds, and a single loss ratio; wherein the target loss rate calculation formula includes:
[0098]
[0099] Among them, n2 represents the expected number of rounds, Indicates the single battle loss ratio. At the same time, the above embodiment provides an example of accumulating the phase battle loss rate to calculate the target battle loss rate. The target battle loss rate can also be calculated by this formula first and then the battle loss rate of each phase can be obtained by splitting it according to the weight of each phase.
[0100] Since different games have different single-battle damage ratios, this patent proposes a method for calculating the target damage ratio based on the expected number of attacks. This method sets the expected number of attacks and adaptively adapts the single-battle damage ratio of the current game to obtain a suitable target damage ratio. The formula evaluates a reasonable target damage ratio by combining the expected number of rounds and the single-battle damage ratio. This process can be adjusted dynamically to cope with different game scenarios. Multiple variables are precisely combined through mathematical formulas to provide accurate target damage ratio predictions. At the same time, based on the processing of the weights of each variable, the system can quickly adapt to different game requirements and plans. Since the provided method can perform subtle control and adjustments to each cycle or stage, the accuracy and effectiveness of decision-making are enhanced.
[0101] Through the above embodiments and examples, the provided method can not only dynamically calculate the target combat loss rate based on the aforementioned data, but also ensure accurate adaptation to different game situations, providing strong support for enhancing the autonomous decision-making performance of the intelligent body.
[0102] The provided method obtains and analyzes the information of the game characters to be played before the start of the simulation game. This information includes: the number of game characters. The combat power of the game characters (their strength level, ability, etc.). Based on the obtained number and combat power information of the game characters, the system generates a corresponding number of agents and assigns appropriate combat power to each agent. This step ensures that the generation of the agent is consistent with the current state of the game environment and supports the rationality of subsequent simulations. The initial configuration of the agent is completed by adding the generated multiple agents to the simulation game scene. After the preparation phase is over, the game of the simulation game can be launched. Using the number and combat power of the agents and game characters, the target combat loss rate for this game is calculated and set. The target combat loss rate is a standard for evaluating the game results and the performance of the agents. In the simulation game, the current combat loss rate of the game characters is obtained in real time. The actual combat loss rate is compared with the pre-set target combat loss rate. If the current combat loss rate exceeds the target combat loss rate, the agent will take a "surrender" behavior to terminate the game. This step ensures that the battle results are within a controllable range to reduce unnecessary waste of resources.
[0103] The proposed method has at least the following beneficial effects:
[0104] 1. Improved flexibility: The method dynamically adjusts the behavior and number of agents according to the real-time situation, overcoming the static limitations of the behavior tree and improving the flexibility and adaptability of the system.
[0105] 2. Complexity management: Adopting decision logic guided by combat damage rate avoids complex behavior tree structure, making subsequent maintenance and expansion easier.
[0106] 3. Risk control: By monitoring the loss rate, the agent can automatically terminate the game when the risk is too high, reducing the exposure risk caused by bad strategies.
[0107] 4. Resource optimization: The intelligent agent makes behavioral decisions based on reasonable calculation and configuration, which improves resource utilization and the overall efficiency of the system.
[0108] In summary, the provided method can form a variety of battle rhythms according to the settings. While meeting the flexibility and dynamics of the AI-trained agent, it is lighter and easier to maintain than the behavior tree as a whole. At the same time, it can make the AI agent's strategy more humane in scenes such as SLG stronghold competition, reduce the risk of AI exposure, and dynamically adapt to the combat power of human players to ensure that the enthusiasm of human players is not dampened while improving the gaming experience of human players. It can also dynamically and adaptively adapt the business's adjustment of the battle loss ratio without retesting the parameters of the adaptation settings. The provided method breaks through the inherent limitations of the behavior tree and manages the agent behavior in a more dynamic and intelligent way, thereby improving the operating efficiency of the simulation game and the performance of the agent.
[0109] See also Figure 4 As shown, Figure 4 1 is a schematic diagram of the structure of the intelligent body control device 200 provided in an embodiment of the present application. The intelligent body control device 200 is used to execute the steps of the intelligent body control method shown in the above embodiments. The intelligent body control device 200 can be a single server or a server cluster, or the intelligent body control device 200 can be a terminal, which can be a handheld terminal, a laptop computer, a wearable device or a robot, etc.
[0110] like Figure 4 As shown, the intelligent body control device 200 includes:
[0111] The information acquisition module 201 is used to acquire the game character information to be played in the simulated game, where the game character information at least includes the number of game characters and the combat power of the game characters;
[0112] A combat power generation module 202, used to generate the number of intelligent agents and the combat power of the intelligent agents according to the number of game characters and the combat power of the game characters;
[0113] An agent generation module 203 is used to generate multiple agents according to the number of agents and the combat power of the agents, and add the generated agents to the simulation game to start the game of the simulation game;
[0114] A target generation module 204, for generating a target combat loss rate according to the number of agents, the combat power of the agents, the number of game characters, and the combat power of the game characters;
[0115] The battle damage acquisition module 205 is used to obtain the current battle damage rate corresponding to the game character in the game of the simulation game;
[0116] The control completion module 206 is used to control the intelligent agent to surrender in the game of the simulation game if the current loss rate is greater than the target loss rate.
[0117] In some embodiments, the target loss rate includes a stage loss rate and a total loss rate, and the total loss rate is greater than the stage loss rate; the control completion module 206 is also used to: if the current loss rate is less than the stage loss rate, update the number of game characters, the combat power of the game characters, the number of intelligent agents and the combat power of the intelligent agents; adjust the number of intelligent agents and the combat power of the intelligent agents according to the updated number of game characters, the combat power of the game characters, the number of intelligent agents and the combat power of the intelligent agents, and restart the game of the simulated game; update the current loss rate until the current loss rate is greater than the stage loss rate; if the current loss rate is greater than the total loss rate, control the intelligent agent to surrender in the game of the simulated game.
[0118] Exemplarily, the control completion module 206 is further used to: generate a game character combat power value according to the number of game characters and the combat power of the game characters; generate an intelligent agent combat power value according to the number of intelligent agents and the combat power of the intelligent agents; obtain the current number of game rounds and the expected number of game rounds of the simulation game; adjust the number of intelligent agents and the intelligent agent combat power according to the game character combat power value, the current number of game rounds and the expected number of game rounds of the intelligent agent combat power value, and the corresponding expression for adjusting the number of intelligent agents and the intelligent agent combat power includes:
[0119]
[0120] Among them, F AI,control Indicates adjusting the combat power of the intelligent body, N AI,control represents the number of agents after adjustment, P AI，control Indicates the adjusted combat power of the intelligent agent, F AI,max Indicates the combat power of the intelligent agent, F u,max Indicates the game character's combat power value, n1 indicates the current round number, and n2 indicates the expected round number.
[0121] Exemplarily, if the current loss rate is less than or equal to the total loss rate, the control completion module 206 is also used to: update the number of intelligent agents, the combat power of the intelligent agents, the number of game characters and the combat power of the game characters; update the stage loss rate according to the current loss rate, the number of intelligent agents, the combat power of the intelligent agents, the number of game characters and the combat power of the game characters, and restart the simulation game until the current loss rate is greater than the stage loss rate and the total loss rate.
[0122] In some embodiments, the target generation module 204 is also used to: obtain the expected number of rounds of the simulation game; generate the corresponding single loss ratio of the simulation game according to the number of intelligent agents, the combat power of the intelligent agents, the number of game characters and the combat power of the game characters; and obtain the target loss rate according to the expected number of rounds and the single loss ratio.
[0123] Exemplarily, the target generation module 204 is further used to calculate the target loss rate according to a preset target loss rate calculation formula, the expected number of rounds and the single loss ratio; wherein the target loss rate calculation formula includes:
[0124]
[0125] Among them, n2 represents the expected number of rounds, Indicates the single battle damage ratio.
[0126] In some embodiments, the root combat power generation module 202 is also used to: obtain difficulty information corresponding to the simulation game; obtain the quantity generation coefficient and combat power generation coefficient corresponding to the intelligent body according to the difficulty information; generate the number of intelligent bodies according to the number of game characters and the quantity generation coefficient; generate the combat power of the intelligent body according to the combat power generation coefficient and the combat power of the game characters.
[0127] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working processes of the intelligent body control device and each module described above can refer to the corresponding processes in the intelligent body control method embodiments described in the above embodiments, and will not be repeated here.
[0128] The above-mentioned intelligent agent control method can be implemented in the form of a computer program. The computer program can be used in Figure 4 Run on the device shown.
[0129] See also Figure 5 , Figure 5 1 is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device includes a processor, a memory and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.
[0130] The storage medium can store an operating device and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any intelligent agent control method.
[0131] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.
[0132] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any intelligent agent control method.
[0133] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the terminal to which the scheme of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0134] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0135] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:
[0136] Obtaining game character information to be played in the simulated game, where the game character information at least includes the number of game characters and the combat power of the game characters;
[0137] Generate the number of intelligent entities and their combat power according to the number of game characters and their combat power;
[0138] Generate multiple agents according to the number of agents and the combat power of the agents, and add the generated agents to the simulation game to start the game of the simulation game;
[0139] Generate a target damage rate based on the number of agents, agent combat power, number of game characters, and game character combat power;
[0140] Get the current combat loss rate of the game characters in the simulated game;
[0141] If the current loss rate is greater than the target loss rate, the agent is controlled to surrender in the simulated game.
[0142] In some embodiments, the target loss rate includes a stage loss rate and a total loss rate, and the total loss rate is greater than the stage loss rate; if the current loss rate is greater than the target loss rate, the intelligent agent is controlled to surrender in the game of the simulation game, including: if the current loss rate is less than the stage loss rate, updating the number of game characters, the combat power of the game characters, the number of intelligent agents, and the combat power of the intelligent agents; adjusting the number of intelligent agents and the combat power of the intelligent agents according to the updated number of game characters, the combat power of the game characters, the number of intelligent agents, and the combat power of the intelligent agents, and restarting the game of the simulation game; updating the current loss rate until the current loss rate is greater than the stage loss rate; if the current loss rate is greater than the total loss rate, controlling the intelligent agent to surrender in the game of the simulation game.
[0143] Exemplarily, adjusting the number of agents and the combat power of agents according to the updated number of game characters, the combat power of game characters, the number of agents and the combat power of agents includes: generating a combat power value of game characters according to the number of game characters and the combat power of game characters; generating an agent combat power value according to the number of agents and the combat power of agents; obtaining the current number of rounds and the expected number of rounds of the simulated game; adjusting the number of agents and the combat power of agents according to the current number of rounds and the expected number of rounds of the game according to the combat power value of the game characters and the combat power value of the agents, and the expressions corresponding to the adjustment of the number of agents and the combat power of agents include:
[0144]
[0145] Among them, F AI,control Indicates adjusting the combat power of the intelligent body, N AI,control represents the number of agents after adjustment, P AI，control Indicates the adjusted combat power of the intelligent agent, F AI,max Indicates the combat power of the intelligent agent, F u,max Indicates the game character's combat power value, n1 indicates the current round number, and n2 indicates the expected round number.
[0146] Exemplarily, if the current loss rate is less than or equal to the total loss rate, the method further includes: updating the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters; updating the stage loss rate according to the current loss rate, the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters, and restarting the simulation game until the current loss rate is greater than the stage loss rate and the total loss rate.
[0147] In some embodiments, a target loss rate is generated based on the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters, including: obtaining the expected number of rounds of the simulated game; generating a corresponding single loss ratio of the simulated game based on the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters; and obtaining the target loss rate based on the expected number of rounds and the single loss ratio.
[0148] Exemplarily, obtaining a target loss rate according to an expected number of game rounds and a single loss ratio includes: calculating the target loss rate according to a preset target loss rate calculation formula, the expected number of game rounds, and a single loss ratio; wherein the target loss rate calculation formula includes:
[0149]
[0150] Among them, n2 represents the expected number of rounds, Indicates the single battle damage ratio.
[0151] In some embodiments, the number of intelligent agents and their combat power are generated according to the number of game characters and their combat power, including: obtaining difficulty information corresponding to the simulation game; obtaining the number generation coefficient and the combat power generation coefficient corresponding to the intelligent agent according to the difficulty information; generating the number of intelligent agents according to the number of game characters and the number generation coefficient; generating the combat power of the intelligent agents according to the combat power generation coefficient and the combat power of the game characters.
[0152] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement the steps of the intelligent body control method provided in the above-mentioned embodiments of the present application.
[0153] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.
[0154] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.
Claims
1. An intelligent agent control method, characterized in that: The method comprises: Obtaining game character information to be played in the simulated game, wherein the game character information includes at least the number of game characters and the combat power of the game characters; Generate the number of intelligent entities and the combat power of the intelligent entities according to the number of game characters and the combat power of the game characters; generating a plurality of intelligent agents according to the number of intelligent agents and the combat power of the intelligent agents, and adding the generated intelligent agents to the simulation game to start a game of the simulation game; Generating a target combat loss rate according to the number of agents, the combat power of the agents, the number of game characters, and the combat power of the game characters; Obtaining a current combat loss rate corresponding to a game character in a match of the simulation game; If the current loss rate is greater than the target loss rate, the agent is controlled to surrender in the game of the simulation game.
2. The method according to claim 1, characterized in that The target loss rate includes a stage loss rate and a total loss rate, and the total loss rate is greater than the stage loss rate; if the current loss rate is greater than the target loss rate, controlling the agent to surrender in the game of the simulation game, comprises: If the current loss rate is less than the stage loss rate, update the number of game characters, the combat power of game characters, the number of intelligent agents, and the combat power of intelligent agents; Adjusting the number of agents and the combat power of agents according to the updated number of game characters, the combat power of game characters, the number of agents and the combat power of agents, and restarting the game of the simulation game; Updating the current battle loss rate until the current battle loss rate is greater than the stage battle loss rate; If the current loss rate is greater than the total loss rate, the agent is controlled to surrender in the game of the simulation game.
3. The method according to claim 2, characterized in that The adjusting the number of agents and the combat power of agents according to the updated number of game characters, the combat power of game characters, the number of agents and the combat power of agents comprises: Generate a game character combat power value according to the number of game characters and the combat power of the game characters; Generate an intelligent agent combat power value according to the number of intelligent agents and the intelligent agent combat power; Obtaining the current number of game rounds and the expected number of game rounds of the simulated game; According to the game character combat power value, the current round number and the expected round number of the intelligent agent combat power value, the number of intelligent agents and the intelligent agent combat power are adjusted. The corresponding expressions for adjusting the number of intelligent agents and the intelligent agent combat power include: Among them, F AI,control Indicates adjusting the combat power of the intelligent body, N AI,control represents the adjusted number of agents, P AI，control represents the adjusted combat power of the intelligent agent, F AI,max Indicates the combat power value of the intelligent agent, F u,max represents the combat power value of the game character, n1 represents the current round number of the game, and n2 represents the expected round number of the game.
4. The method according to claim 2, characterized in that: If the current battle loss rate is less than or equal to the total battle loss rate, the method further includes: Updating the number of agents, agent combat power, number of game characters, and game character combat power; The stage loss rate is updated according to the current loss rate, the number of intelligent agents, the combat power of the intelligent agents, the number of game characters and the combat power of the game characters, and the simulation game is restarted until the current loss rate is greater than the stage loss rate and the total loss rate.
5. The method according to claim 1, characterized in that The generating a target combat loss rate according to the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters includes: Obtaining an expected number of rounds of the simulated game; Generate a single battle loss ratio corresponding to the simulation game according to the number of intelligent agents, the combat power of the intelligent agents, the number of game characters, and the combat power of the game characters; The target loss rate is obtained according to the expected number of game rounds and the single loss ratio.
6. The method according to claim 5, characterized in that The obtaining the target battle loss rate according to the expected number of game rounds and the single battle loss ratio includes: Calculating the target loss rate according to a preset target loss rate calculation formula, the expected number of rounds and the single loss ratio; The target damage rate calculation formula includes: Wherein, n2 represents the expected number of rounds, Indicates the single combat damage ratio.
7. The method according to claim 1, characterized in that The generating the number of intelligent agents and the combat power of the intelligent agents according to the number of game characters and the combat power of the game characters comprises: Obtaining difficulty information corresponding to the simulation game; Obtaining a quantity generation coefficient and a combat power generation coefficient corresponding to the intelligent agent according to the difficulty information; Generate the number of agents according to the number of game characters and the number generation coefficient; The intelligent body combat power is generated according to the combat power generation coefficient and the game character combat power.
8. An intelligent agent control device, characterized in that: include: An information acquisition module, used to acquire information about game characters to be played in the simulated game, wherein the game character information at least includes the number of game characters and the combat power of the game characters; A combat power generation module, used to generate the number of intelligent entities and the combat power of the intelligent entities according to the number of game characters and the combat power of the game characters; An agent generation module, used for generating a plurality of agents according to the number of agents and the combat power of the agents, and adding the generated agents to the simulation game to start a game of the simulation game; A target generation module, used for generating a target combat loss rate according to the number of intelligent agents, the combat power of the intelligent agents, the number of game characters and the combat power of the game characters; A battle damage acquisition module, used to acquire the current battle damage rate corresponding to the game character in the game of the simulation game; The control completion module is used to control the agent to surrender in the game of the simulation game if the current loss rate is greater than the target loss rate.
9. A computer device, characterized in that: The computer device comprises: Memory and processor; Wherein, the memory is connected to the processor and is used to store programs; The processor is configured to implement the steps of the method according to any one of claims 1 to 7 by running the program stored in the memory.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, enables the processor to implement the steps of the method according to any one of claims 1 to 7.