Competitive game processing method and apparatus, electronic device and storage medium

Abstract

A competitive game processing method and apparatus, an electronic device and a storage medium. The method comprises: in response to a first player selecting a first competitive mode, selecting from a game session set a target game session corresponding to the first player, wherein the game session set comprises a player versus player (PVP) session and a plurality of different types of player versus environment (PVE) sessions; in response to the target game session being a PVE session, starting a game corresponding to the target game session for the first player; and during the gameplay runtime of the target game session, dynamically adjusting difficulty parameters of robots in the target game session, and controlling the robots to carry out game behaviors according to the adjusted difficulty parameters until the game of the target game session is finished, the difficulty parameters affecting the intensity of the game of the target game session and / or the game result of the first player at the end of the game of the target game session. The technique of the present disclosure can improve player's flow experience in gaming, thus reducing the player churn rate.

Description

Method, device, electronic device and storage medium for processing battle games

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims priority to Chinese patent application number 202410107392.5, filed on January 25, 2024, entitled “Method, device, electronic device and storage medium for processing battle games”, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to the field of game technology, and in particular to a method, device, electronic device, and storage medium for processing a battle game. Background Art

[0004] Competitive games typically support both real-person and bot battles. Before the game begins, the system will match players with either a real person or a bot based on the actual situation.

[0005] In human-robot combat scenarios, traditional robots typically use pre-designed behavior trees to determine their gameplay. Behavior trees make decisions based on predefined behavior nodes and rules. For complex game situations and multiple behavioral options, the behavior tree hierarchy can become complex. Furthermore, this approach may not adapt to the dynamics and uncertainties of the game when dealing with varying player behaviors and game situations. Consequently, in robot combat scenarios based on behavior trees, their gameplay is relatively simple, making it easy for human players to find patterns in the game. This results in a poor flow experience for human players and increases their chances of churn. Summary of the Invention

[0006] The purpose of the present disclosure is to provide a method, device, electronic device and storage medium for processing battle games to enhance the player's gaming flow experience and reduce player churn.

[0007] An embodiment of the present disclosure provides a method for processing a battle game, the method comprising: in response to a first player selecting a first competitive mode, selecting a target game corresponding to the first player from a game game set; wherein the game game set includes a real-person game and multiple different types of human-computer games; in response to the target game being a human-computer game, opening a game corresponding to the target game for the first player; during the game of the target game, dynamically adjusting the difficulty parameters of a robot in the target game, and controlling the robot to perform gaming behavior according to the adjusted difficulty parameters until the game of the target game ends; wherein the difficulty parameters affect the intensity of the game of the target game and / or the win / loss result of the first player at the end of the game of the target game.

[0008] In a second aspect, the disclosed embodiment further provides a device for processing a battle game, the device comprising: a target round selection module, configured to execute, in response to a first player selecting a first competitive mode, selecting a target round corresponding to the first player from a game round set; wherein the game round set includes a real-person round and multiple different types of human-machine rounds; a game start module, configured to execute, in response to the target round being a human-machine round, starting a game corresponding to the target round for the first player; a parameter adjustment module, configured to execute, during the course of the game of the target round, dynamically adjusting the difficulty parameters of the robot in the target round, and controlling the robot to perform game behavior according to the adjusted difficulty parameters until the game of the target round ends; wherein the difficulty parameters affect the intensity of the game of the target round and / or the win or loss result of the first player at the end of the game of the target round.

[0009] On the third aspect, an embodiment of the present disclosure further provides an electronic device, including a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the above-mentioned processing method of the fighting game.

[0010] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the above-mentioned processing method of the battle game.

[0011] The disclosed embodiments provide a method, device, electronic device, and storage medium for processing a battle game. After a player selects a competitive mode, the player selects a target game from a game game set including a real-person game and multiple different types of human-computer games. This allows the target game to no longer be a single type of game game, but to have a richer variety of forms, thereby allowing the player to experience greater differences in multiple consecutive game experiences. When the target game is a human-computer game, the difficulty parameters of the robot can be controlled during the game, thereby affecting the intensity of the game and / or the winning or losing outcome of the first player at the end of the target game. At the same time, the game behaviors executed by the robot are made more diverse, reducing the monotony of the game caused by the robot's routine game behaviors, allowing the player to enjoy a game process similar to that of other real-person player games, thereby enhancing the player's flow experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the related technologies, the following briefly introduces the drawings required for use in the specific embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0013] FIG1 is a flowchart of a processing method of a battle game provided by one embodiment of the present disclosure;

[0014] FIG2 is a framework diagram of a battle game processing system provided by one embodiment of the present disclosure;

[0015] FIG3 is a schematic diagram of the structure of a processing device for a battle game provided by one embodiment of the present disclosure;

[0016] FIG4 is a schematic diagram of the structure of another battle game processing device provided by one embodiment of the present disclosure;

[0017] FIG5 is a schematic diagram of the structure of another battle game processing device provided by one embodiment of the present disclosure;

[0018] FIG6 is a schematic structural diagram of an electronic device provided by one embodiment of the present disclosure. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions of the present disclosure in conjunction with the embodiments. Obviously, the embodiments described are only a part of the embodiments of the present disclosure, not all of them. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present disclosure.

[0020] The flow experience of a game can be understood as an immersive state in which players continuously gain positive, pleasant and fulfilling feelings when they highly devote their energy to game activities. During the game process, there are many factors that affect the player's flow experience. For example: the issue of game win rate. The core flow experience of players is usually to win the game. If a game match is lost, or if several consecutive game matches are lost, the player's flow experience will be reduced. However, if consecutive basic game matches are won, the player will also feel that the game is not challenging, thereby reducing the flow experience. In addition to the game win rate, there are some indicators, such as the length of the match, that is, the waiting time before entering the game, which will also affect the player's flow experience. In addition, for example, if a real player encounters a player who uses script plug-ins, it will also affect the real player's flow experience. In order to enhance the player's game flow experience, the embodiments of the present disclosure provide a processing method, device, electronic device and storage medium for a battle game.

[0021] This embodiment provides a method for processing a competitive game. The method can be run on a terminal device or a server. The terminal device can be a local terminal device. When the method for processing a competitive game is run on a server, the method can be implemented and executed based on a cloud interaction system, wherein the cloud interaction system includes a server and a client device.

[0022] In an optional embodiment, various cloud applications, such as cloud games, can be run under the cloud interaction system. Taking cloud games as an example, cloud games refer to a gaming method based on cloud computing. In the cloud gaming operation mode, the operating body of the game program and the main body of the game screen presentation are separated. The storage and operation of the processing method of the battle game are completed on the cloud gaming server. The role of the client device is to receive and send data and present the game screen. For example, the client device can be a display device with data transmission function close to the user side, such as a mobile terminal, TV, computer, PDA, etc.; but the terminal device for information processing is the cloud gaming server in the cloud. When playing the game, the player operates the client device to send operation instructions to the cloud gaming server. The cloud gaming server runs the game according to the operation instructions, encodes and compresses the game screen and other data, and returns it to the client device through the network. Finally, the client device decodes and outputs the game screen.

[0023] In an optional embodiment, the terminal device can be a local terminal device. Taking a game as an example, the local terminal device stores the game program and is used to present the game screen. The local terminal device is used to interact with the player through a graphical user interface, that is, conventionally downloading and installing the game program through an electronic device and running it. The local terminal device can provide the graphical user interface to the player in various ways, for example, it can be rendered and displayed on the terminal display, or provided to the player through holographic projection. For example, the local terminal device may include a display screen and a processor, the display screen is used to present the graphical user interface, the graphical user interface including the game screen, and the processor is used to run the game, generate the graphical user interface, and control the display of the graphical user interface on the display screen.

[0024] In one possible implementation, the disclosed embodiments provide a method for processing a battle game, providing a graphical user interface via a terminal device, wherein the terminal device can be the local terminal device mentioned above or a client device in the cloud interaction system mentioned above. A graphical user interface is provided via the terminal device, the graphical user interface including at least a portion of a game scene, a first virtual object with a target identity, and multiple second virtual objects in a live state, wherein the multiple second virtual objects are other virtual objects in the same game as the first virtual object. As described in the aforementioned embodiment, the first virtual object with a target identity has the authority to attack other virtual objects. The attacked virtual object enters a defeated state and no longer participates in subsequent game progress. The target identity is an identity attribute assigned at the start of the game. Among the multiple virtual objects participating in the game, there can be one or more robots, that is, the game is a human-computer game, and the virtual object corresponding to the robot can be either a teammate of the first virtual object or an enemy of the first virtual object. The robot (also called an intelligent agent or AI agent) in this embodiment is a pre-trained robot model that can perform game behaviors according to the actual game scene in the game.

[0025] This embodiment uses RL (deep reinforcement learning) technology or BC (behavior cloning) technology to train the robot. During the specific training process, player data can be collected for imitation learning, or a pre-designed reward function can be used to enable the robot model to interactively learn in a real game through reinforcement learning, ultimately obtaining a robot whose behavior meets expectations. For example, the robot's win rate, the ratio between the release probability of certain specific actions and the release probability of virtual objects operated by real people, etc. meet expectations.

[0026] Referring to the flowchart of a method for processing a battle game shown in FIG1 , the method can be applied to the above-mentioned terminal device and specifically includes the following steps:

[0027] Step S102 , in response to the first player selecting the first competition mode, selecting a target game corresponding to the first player from a game game set; wherein the game game set includes a real-person game and multiple different types of human-computer games.

[0028] The first competitive mode mentioned above is a type of multiplayer game. There are different gameplay modes with different rules under the same game. Each gameplay mode is regarded as a competitive mode, such as melee mode, team battle mode, one-on-one mode, three-on-three mode, etc.

[0029] The game rounds in the aforementioned game round set can be pre-configured based on actual application needs. Within the same game and the same competitive mode, different game rounds may have different combat rhythms, win rates, and other factors. The player's choice of game round type is not determined by the player but by the game system. As one possible implementation, the target round can be determined based on at least one of the following information: the first player's gaming skill, historical win-loss rate, and performance in the previous game.

[0030] The player objects in the aforementioned live games are all virtual objects corresponding to actual users. At least one player object in the aforementioned human-machine games is a virtual object corresponding to a robot. Human-machine games can include multiple different types, with different robots in each type. For example, different types of human-machine games can be divided based on the win rate provided to the first player, or based on the number of robots included in the game, or even based on the combat style of the robots in the game.

[0031] Step S104 , in response to the target round being a human-machine round, opening a game corresponding to the target round for the first player.

[0032] Opening the game corresponding to the target round for the first player includes combining real players and robots so that the number of virtual characters in the target round meets the minimum number of players required for the game; and placing each virtual character at a corresponding birthplace in the game scene.

[0033] Step S106, during the target round of the game, dynamically adjust the difficulty parameters of the robots in the target round, and control the robots to perform game behaviors according to the adjusted difficulty parameters until the target round of the game ends; wherein the difficulty parameters affect the intensity of the game in the target round and / or the winning or losing result of the first player at the end of the target round of the game.

[0034] The difficulty parameter of the robot directly reflects information about its combat reaction ability, reasoning ability, and skill proficiency. By adjusting the difficulty parameter of the robot during the game, the game behavior performed by the robot can be made more diverse and more similar to the corresponding game behavior of a real person controlling a virtual character.

[0035] The above method of this embodiment, after the player selects the competition mode, selects the target game from a game game set including real-person games and multiple different types of human-computer games, so that the target game is no longer a single type of game game, but has a richer variety of forms, thereby making the player's gaming experience more diverse over multiple consecutive games; when the target game is a human-computer game, the difficulty parameters of the robot can be controlled during the game, thereby affecting the intensity of the game and / or the winning or losing outcome of the first player at the end of the target game, and at the same time making the gaming behavior executed by the robot more diverse, reducing the monotony of the game caused by the routine gaming behavior of the robot, allowing the player to enjoy a gaming process similar to that of other real-person player games, thereby enhancing the player's flow experience.

[0036] As one possible implementation, selecting a target game corresponding to the first player from the game game set includes: obtaining the first player's game history data; wherein the game history data includes game skill data (e.g., the first player's rank) and / or win / loss data for the last N game games, where N is a positive integer; and selecting the target game corresponding to the first player from the game game set based on the game history data. By combining the first player's game history data, the target game selected can better suit the first player's actual situation, making the target game more engaging for the first player and enhancing the first player's flow experience.

[0037] The selection of target games can be accomplished by a pre-trained AI matching model, which offers greater flexibility and versatility than fixed matching strategies. The AI ​​matching model abstracts the in-game matching rules and converts them into rule templates. By flexibly combining these rule templates, matching rules tailored to specific business needs can be designed. Different games have different rule templates. For example, in Game A, the rank difference between the two factions must not exceed 1, while in Game B, which doesn't have a rank concept but does have an equipment rating, the difference must not exceed 10,000. These are two distinct rules for the game. The AI ​​matching model can abstract them into a rule template: xy<=M. This single rule template can then be applied to both Game A and Game B, requiring only x, y, and M to be modified based on the actual game rules. In selecting a target game using an AI matching model, game history data may include at least one of the following: the first player's historical win rate, performance data from the previous game, and the first player's acceptance of human-computer games. This performance data may include player operational capabilities (such as player ability scores or skill proficiency), game path data, historical scores (number of wins in matches, data on enemy character hits and hits by enemies), and game preference data (such as which hero character types the player prefers). Player acceptance of human-computer games can be determined through in-game research and experience. Generally, mid- and low-level players have a higher acceptance of human-computer games, while high-level players have a lower acceptance. Based on this, selecting a target game corresponding to the first player from a set of game games based on game history data includes: inputting the game history data into a pre-trained AI matching model to obtain a human-computer game profit value for the first player; and selecting a target game corresponding to the first player from the set of game games based on the human-computer game profit value. The human-computer game profit value refers to a value obtained by comparing the predicted profit value of the corresponding live game. For example: if the predicted score of the player in the real-person game is 0.5, and the predicted score of the human-computer game is 0.6, then the profit value of the human-computer game is 0.1. As a possible implementation method, the AI ​​matching model can give the profit value of each type of human-computer game, and then determine the target game based on the profit value of each type of human-computer game. Alternatively, when the profit value of the human-computer game reaches or exceeds the set profit threshold, the target game can be determined from the human-computer game, otherwise the real-person game is determined as the target game, and the first player is matched with a real-person player to play the game. By using the AI ​​matching model to select the target game based on the profit value of the human-computer game, the profit of the first player in the target game can be maximized, thereby making the first player have a higher sense of accomplishment.

[0038] Based on the above AI matching model, the specific selection process of the target game can also be screened layer by layer, for example as follows:

[0039] (1) Rule strategy (combination optimization): used to set the matching rule strategy for the game. This rule strategy is a relatively hard restriction and must be met in principle.

[0040] (2) Optimization goal (combination optimization): When the matching rule strategy is satisfied, further search for a better solution (combination). This optimization goal is a soft constraint and does not have to be satisfied, but it is better to satisfy it.

[0041] (3) Prediction Model (AI): Under the above matching rule strategy, the AI ​​matching model is used to further find the optimal solution (combination) from the optimization target result. The AI ​​matching model can make predictions based on the player's historical win rate, performance in recent games, player skill level, playing style, etc., and then determine the profit value corresponding to the target game. The game corresponding to the maximum profit value is selected as the target game.

[0042] As a possible implementation, some players may prefer to play in a live-action game. Based on this, selecting a target game corresponding to the first player from the game game set based on the game history data includes: matching the first player with a live-action player corresponding to the live-action game based on the game history data; and in response to a failure to match the live-action game within a first set time period, selecting a target game corresponding to the first player from the human-machine game set. In this way, the first player can prioritize playing in a live-action game. If a live-action game fails to match (e.g., at night or during a certain period of working hours, there are no live players who have selected the first competitive mode, or there are insufficient such live players), then a target game is selected from the human-machine game set. Because the robots in this embodiment are more diverse, such a human-machine game is difficult for players to detect as a battle against robots, which can also meet the player's gaming needs and reduce player waiting time.

[0043] As a possible implementation method, in order to further enhance the gaming interest of mid- and low-level players, the aforementioned game history data includes game level data (such as rank or skill level, etc.); based on this, the aforementioned selection of a target game corresponding to the first player from the game game set based on the game history data includes: if the aforementioned game level data indicates that the first player's gaming level is higher than a set level threshold, selecting a real-person game from the game game set as the target game corresponding to the first player, and matching the first player with a real-person player corresponding to the real-person game based on the game level data; if the aforementioned game level data indicates that the first player's gaming level is not higher than the set level threshold, selecting a target game corresponding to the first player from the human-computer game in the game game set. By selecting a real-person game as the target game for high-level players, the gaming competitive ability of high-level players can be more fully utilized, thereby enhancing the flow experience of high-level players; by selecting a human-computer game as the target game for low-level players, the robot can provide more appropriate difficulty parameters during the game, maintaining the player's win-loss rate within a reasonable range, thereby enhancing the flow experience of low-level players.

[0044] In order to keep the player's win-loss rate within a reasonable range, the game history data may also include the win-loss data of the last N games; selecting the target game corresponding to the first player from the human-computer game set includes: (1) determining the expected result of the current game based on the win-loss data of the last N games, for example, it can be predicted by the AI ​​matching model, or it can be predicted based on the win-loss data of the last N games and the reasonable range of the win-loss rate corresponding to the first player's current level. (2) selecting the target game corresponding to the first player from the human-computer game set based on the expected result. For example: the win-loss rate of players in the middle level is controlled in the range of 60-70%, so that these players always have a slight advantage during the game process, thereby increasing their gaming pleasure. If the player loses the first two games, the third game must be won by the player. The player can win by selecting the target game in the third game and controlling the robot. By determining the expected result of the current game based on the win-loss data of the player's last N games and selecting the target game based on the result, the selection of the target game can be made more reasonable and the player's win-loss rate can be maintained within a reasonable range.

[0045] As a possible implementation, selecting the target game corresponding to the first player from the human-machine games in the game game set according to the expected result may include: (1) if the expected result is victory for the first player, selecting the first type of human-machine game from the human-machine games in the game game set as the target game corresponding to the first player; wherein the first type of human-machine game provides a higher winning rate for the first player than other human-machine games; (2) if the expected result is defeat for the first player, selecting the second type of human-machine game from the human-machine games in the game game set as the target game corresponding to the first player; wherein the second type of human-machine game provides a lower winning rate for the first player than other human-machine games. The winning rate provided by the human-machine game to the player can be mainly controlled by the type and number of robots configured in the game. For example, more robots can be configured in the first type of human-machine game. Taking a 3-on-3 competitive mode as an example, 5 robots can be configured in the first type of human-machine game, with only one real person, the first player. By adjusting the difficulty parameters of the 5 robots, the first player can be guaranteed to win. In the second type of human-machine game, three robots can be deployed: the first player and two other real players. Two lower-level real players can also be deployed to form a team with the first player. This way, by controlling the difficulty parameters of the three robots, the first player can be prevented from losing the game. Robot types can also be differentiated. During robot training, robots of different styles can be trained, such as strong and weak robots, and the corresponding robot types can be configured for the human-machine game as needed. This method, by selecting the first or second type of human-machine game as the target game based on the first player's win-loss information in the expected outcome, ensures that the first player achieves the expected result in the target game.

[0046] To make the player's gaming experience feel more realistic, some milder human-machine games can be interspersed between winning and losing games, so that the player's win or loss in this game has a certain degree of randomness. Based on this, the above-mentioned game history data can also include the type of the previous game; selecting the target game corresponding to the first player from the human-machine games in the game game set includes: if the type of the previous game is the first type of human-machine game or the second type of human-machine game, selecting the third type of human-machine game from the human-machine games in the game game set as the target game corresponding to the first player; wherein the first type of human-machine game provides the first player with a higher win rate than other human-machine games, the second type of human-machine game provides the first player with a lower win rate than other human-machine games, and the third type of human-machine game provides the first player with a win rate between the first type of human-machine game and the second type of human-machine game. By determining the target game of the current game based on the type of the previous game, the types of adjacent games can be more diverse, improving the player's gaming flow experience.

[0047] For some scenarios, it can be recommended to prioritize selecting a real-person game as the target game for the player. In order to avoid player loss due to waiting too long for a game to be formed, the above-mentioned selection of the target game corresponding to the first player from the game game set may include: selecting a real-person game as the target game corresponding to the first player from the game game set, and matching the first player with a real-person player corresponding to the real-person game; in response to the first player waiting for a second set time (the time can be set based on experience) and not matching a real-person player in the real-person game, selecting a fourth type of human-computer game from the human-computer games in the game game set as the target game corresponding to the first player; wherein the fourth type of human-computer game provides a random value for the winning rate of the first player, and the game may include only the first player, a real-person player, and the rest of the players are robots. By promptly providing the first player with the fourth type of human-computer game for playing when the time for not matching a real-person player in the real-person game reaches the second set time, the situation where the first player loses patience and logs off due to continued waiting can be reduced, thereby improving their flow experience.

[0048] In some games, some players may use scripts or other AI models to control their virtual characters. These players are called cheat-type players. When a cheat-type player is present in a team of real players, the real players can clearly perceive that the cheat-type player's gaming performance is relatively slack, causing the team to fail in the game, which in turn affects the real players' flow experience. Based on this, the above method also includes: determining whether the first player is a cheat-type player based on historical game data; if so, selecting a fifth type of human-computer game from the game game set and opening a game corresponding to the fifth type of human-computer game for the first player; wherein the other players in the fifth type of human-computer game are all robots or cheat-type players; if not, executing the above step of selecting the target game corresponding to the first player from the game game set based on historical game data. By setting up this type of game game, cheat-type players can be isolated from real players, preventing cheat-type players from interfering with real players, thereby ensuring the real players' flow experience.

[0049] In order to make the robot's gaming behavior richer, the robot in this embodiment can have different styles, which refers to the style of the robot's own behavioral strategy. For example, the difference between "reckless" or "timid" of a single robot, that is, whether it cares more about defeating the opponent / saving itself, is a style.

[0050] Robots with different styles can be trained. To improve training effectiveness, the training samples can be generated by the interaction between the robot model and the environment during learning. This approach is also called self-imitation learning, meaning that the samples are generated by the model or strategy itself during training. During reinforcement learning, these samples are filtered or selected based on a pre-set style selection rule or reward scheme, ensuring that the remaining samples conform to the pre-set style. For example, for an aggressive style, rules are used to select samples that exhibit intense combat, such as those with less than 5% health at the final victory. By reinforcing these samples, the trained robot ultimately tends to favor that style.

[0051] This reward setting (also known as reward shaping) is a method for adjusting learning objectives in reinforcement learning. The training goal of reinforcement learning is to maximize the expected reward. By increasing the coefficient of the style-related sub-item in the reward, the trained robot can ultimately be biased towards that style.

[0052] In addition, in some fighting games, multiple robots can be trained as a group to form a team-level style (or tactics), similar to the tactical style of basketball games that focuses on outside shooting or inside shooting.

[0053] The difficulty and behavior control of the robot in this embodiment can be adjusted during training by providing adjustable parameters corresponding to the model, rules, and game attributes during the training process. These adjustable parameters include: model version number, model parameter size, model training duration, and different reward functions. The variance of the Gaussian distribution of a perturbation in a certain dimension of the model input and the probability of a certain index being blocked in the model output can also be adjusted.

[0054] These model dimensions can include the differences within the model itself, as well as control over its input and output. Input control involves perturbing or masking specific dimensions, such as adding noise to the opponent's coordinates or health to simulate a human's inaccurate estimation of the opponent's current state. Output control involves masking the "legal" flag indicating whether an action is available, misleading the model into believing it is unavailable, or probabilistically selecting a suboptimal or specific action.

[0055] The aforementioned rules and game attributes are related to the specific game settings. These rules may include player matching rules, gameplay rules, and so on. Game attributes may include attributes related to game characters, virtual weapons, skills, or virtual equipment. Examples include the probability of releasing a specific skill and the training value. This training value varies from game to game. For example, in basketball games, there are three different terms referring to training: potential, training, and talent. These terms, when reflected in the game, will affect a player's shooting accuracy, success rate of driving / stealing, and the performance of special skills.

[0056] The control of the robot's style dimension is to add style feature vectors during the robot's training process, so that the robot has the ability to control style.

[0057] Through the above training methods, the following technical indicators of the robot can be improved:

[0058] (1) Combat strength: combat response ability, reasoning ability, skill proficiency, etc. Through training, the robot can have strong combat strength.

[0059] (2) Behavioral diversity and anthropomorphism: The changes in moves during combat are relatively rich and varied, and can be changed according to the actual game scene. There are no fixed rules and it is not easy to find routines.

[0060] Based on the trained robots, starting the game corresponding to the target round for the first player may include: determining the robot combat style and number of robots in the target round based on the game level data of each human player in the target round and the first competitive method; deploying the robots in the target round for the first player based on the robot combat style and number of robots; and starting the game corresponding to the target round. After determining that the target round is a human-machine round of a certain type, appropriate robots may be assigned based on the skill levels and playing styles of all human players in the target round, such as assigning robots with a defensive style to players with comparable skill levels or who prefer to attack. By determining the robot combat style and number of robots in the target round based on the game level data of the human players and the first competitive method, the robots in the target round can be more closely matched to the game levels of the human players, thereby improving the rationality of the game rhythm.

[0061] To improve the rationality of adjusting the robot's difficulty parameters, the dynamic adjustment of the robot's difficulty parameters in the target round can include: obtaining the first player's performance data in the target round; and adjusting the robot's difficulty parameters in the target round based on this performance data and the type of the target round, so that the first player's win rate in the target round meets the expected win rate. This performance data may include data on skill release proficiency, game route data, data on enemy character hits, and data on being hit by enemies. This expected win rate can be predicted by the AI ​​matching model described above, or it can be found from the system's default expected win rate based on the first player's gaming level. Adjusting the robot's difficulty parameters in this way can make the robot behave more humanlike throughout the game.

[0062] In order to further motivate players, when adjusting the robot difficulty parameters, the difficulty parameters of one or several robots can be adjusted up or down based on the player's reward information (such as game scores or virtual props awarded in the game), so that the battles in the game fluctuate and become more interesting. Based on this, the above adjustment of the difficulty parameters of the robots in the target game based on performance data and the type of the target game includes the following steps (1) and (2):

[0063] (1) determining an adjustment direction for a difficulty parameter of a target robot in the target game based on the reward information in the performance data and the type of the target game; wherein the target robot is a friend or an enemy of the first player, and the difficulty parameter includes at least one of the following: reaction speed, accuracy of judgment of the robot's own or enemy's state, proficiency in skill release, etc.; the adjustment direction includes increasing or decreasing the difficulty parameter;

[0064] The target robot may be determined based on the number of robots included in the target game and the relationship between the robot and the first player.

[0065] The number of robots in a target round varies in different application scenarios. For example, in a three-on-three competitive game, assuming the first player's team is Team 1 and their opponent's team is Team 2, the target rounds can have at least the following formats: Target Round 1: Team 1 consists of three human players, Team 2 consists of three robots; Target Round 2: Team 1 consists of one human player and two robots, Team 2 consists of one human player and two robots; Target Round 3: Team 1 consists of one human player and two robots, Team 2 consists of three robots; Target Round 4: Team 1 consists of three human players, Team 2 consists of one human player and two robots. Different target round formats and different types of target rounds, such as the first and second types of human-machine rounds, will require different target robots and adjustments to their difficulty parameters. Each adjustment can be made to increase the intensity of the competition and add to the dynamics of the game.

[0066] (2) Adjust the difficulty parameters of the target robot according to the determined adjustment direction.

[0067] By adjusting the difficulty parameters of the robot described above, the intensity of the competition or the twists and turns of the competition can be increased, thereby enhancing the fun of the game.

[0068] The aforementioned robot difficulty parameter adjustment process can also be implemented using a pre-trained AI adjustment model. This AI adjustment model can score the abilities of both humans and robots based on the aforementioned data during gameplay, compare the ability scores of both teams, and make adjustments based on the comparison results. For example, in the aforementioned first target round, if the comparison results show a large difference in scores between the first and second teams, the ability scores of the robots in the second team can be lowered to narrow the gap and increase the intensity of the competition.

[0069] The aforementioned ability score can be a single score, a comprehensive evaluation of the in-game abilities of both players and bots. A player's ability score can primarily take into account factors such as the player's win rate, the abilities of their opponents and teammates in each round, and in-game performance. A bot's ability score can be determined based on the bot's data from that particular round. For example, before a bot goes online, a large number of offline bot battles can be conducted to automatically and accurately assess the strength of each bot template, improving the accuracy of human-bot matchmaking. The game can have a number of pairings based on different combinations of bot model versions and difficulty parameters. These different pairings can be matched to different bots, which then play against each other to determine the strength distribution (or, in other words, the win rate and ability score of each bot in each battle). Thus, each bot participating in a target round can have its ability score determined based on its performance in the game's difficulty parameters.

[0070] During the target game, the first player can adjust the bot's difficulty based on changes in skill scores to heighten the intensity of the match. Alternatively, if the score between the two players fluctuates, or the score fluctuates significantly, the bot's difficulty can be adjusted to increase the dynamics of the match and avoid a one-sided score. In certain matches, such as the first type of human-bot match, the human player should be guaranteed to win.

[0071] To reflect the true skill level of each live player during a live match, the system does not intervene during live matches. To collect more accurate player data, after each match, regardless of whether it's live or AI, player skill scores are updated based on their performance, influencing whether the next round will be AI. Based on this, the method further includes: responsive to the conclusion of a target round, recording the first player's game data from the target round; this game data is used to initiate the next game for the first player. This data collection method allows for convenient and reliable data, providing a reference for matchmaking in subsequent rounds.

[0072] Referring to the framework diagram of the battle game processing system shown in FIG2 , after a player logs in to the system, the system offers four competitive modes (or gameplay methods) for the player to choose from. Specifically, the competitive modes are shown in the figure as Modes 1 through 4. The player can select one based on their current needs, such as Mode 1 (3-on-3). Upon receiving the player's selection of Mode 1, the system assigns a game round to the player, for example, matching the player with a corresponding game round based on the player's gaming history. The game rounds provided in this system include: live-action, first-type human-machine, second-type human-machine, third-type human-machine, fourth-type human-machine, and fifth-type human-machine. Live-action games are primarily organized to balance the game proficiency of both teams, and a more flexible live-action matching method can be employed during the specific game organization process. The first-type human-machine game is designed to increase the player's win rate and enhance their enjoyment. The fourth-type human-machine game is designed to minimize player waiting times and allow for faster game formation. The second-type human-machine game primarily provides a certain level of challenge. The third-type human-machine game lies between the first-type and second-type human-machine games, preventing a sudden drop in difficulty between games. The fifth type of human-computer game is a special game mainly for players who use plug-ins.

[0073] If the player is matched with a real person, the system will not intervene during the game.

[0074] If a player is matched with any of the aforementioned human-machine games, a comprehensive assessment of human-machine ability scores can be conducted, allowing the robot and player to be compared across various common dimensions, thereby assigning the appropriate robot to the matching human-machine game. Furthermore, the robot corresponding to the human-machine game can be selected based on the robot's difficulty parameters and style parameters. During the game, the system can also intervene in the robot, effectively acting as an invisible hand, intervening in the robot in the human-machine game to improve the robot's performance in the game. Specifically, the system can adjust the robot's difficulty parameters based on the score difference between the two teams, or further, based on the predicted win rate corresponding to the human-machine game and the player's performance data in the game, to adjust the robot's difficulty parameters.

[0075] If the current game is over, the game data of the players in this game will be collected, such as player behavior data, game win-loss data, level data, etc., to optimize the matching of the next game.

[0076] Through the above method, universal processing of battle games can be achieved. Different games with different playing methods can share the above processing system. By adjusting the matching mechanism of the game in real time, the player experience can be effectively adjusted, and the artificially defined rules and strategies in the game can be coordinated with AI for the matching process of the game. Through the AI ​​model, the flow changes of each player can be quickly identified, and the robot difficulty parameters can be adjusted for each player to enhance the player's flow experience.

[0077] Corresponding to the above method, this embodiment further provides a device for processing a battle game, which can be applied to the above terminal device. Referring to FIG3 , the device includes the following modules:

[0078] The target game selection module 32 is configured to select a target game corresponding to the first player from a game game set in response to the first player selecting the first competition mode; wherein the game game set includes a real-person game and multiple different types of human-computer games;

[0079] The game starting module 34 is configured to execute, in response to the target game being a human-machine game, starting a game corresponding to the target game for the first player;

[0080] The parameter adjustment module 36 is configured to dynamically adjust the difficulty parameters of the robots in the target round during the game, and control the robots to perform game behaviors according to the adjusted difficulty parameters until the game of the target round ends; wherein the difficulty parameters affect the intensity of the game in the target round and / or the winning or losing result of the first player at the end of the game of the target round.

[0081] The above-mentioned symptoms of this embodiment are that after the player selects the competition mode, the target game is selected from a game game set including real-person games and multiple different types of human-computer games, so that the target game is no longer a single type of game game, and the forms are more diverse, thereby making the player's gaming experience more diverse over multiple consecutive games; when the target game is a human-computer game, the difficulty parameters of the robot can be controlled during the game, thereby affecting the intensity of the game and / or the winning or losing outcome of the first player at the end of the target game. At the same time, the gaming behavior executed by the robot is made more diverse, reducing the monotony of the game caused by the routine gaming behavior of the robot, allowing the player to enjoy a gaming process similar to that of other real-person player games, thereby enhancing the player's flow experience.

[0082] As a possible implementation, the target game selection module 32 is further configured to: obtain the first player's game history data; wherein the game history data includes game skill data and / or win / loss data for the last N game games, where N is a positive integer; and select a target game corresponding to the first player from a set of game games based on the game history data. By combining the first player's game history data, the target game selected can better suit the first player's actual situation, making the target game more engaging for the first player and enhancing the first player's flow experience.

[0083] As a possible implementation, the game history data includes at least one of the following: the first player's historical win rate, performance data from the previous game, and the first player's acceptance of human-computer games. Accordingly, the target game selection module 32 is further configured to: input the game history data into a pre-trained AI matching model to obtain the first player's human-computer game profit value; and select a target game corresponding to the first player from a set of game games based on the human-computer game profit value. By using the AI ​​matching model to select the target game based on the human-computer game profit value, the first player's profit in the target game can be maximized, thereby enhancing the first player's sense of accomplishment.

[0084] As a possible implementation, the target game selection module 32 is further configured to execute: matching the first player with a real player corresponding to the real game according to the game history data;

[0085] In response to a failure to match the live game within the first set time period, a target game corresponding to the first player is selected from the human-machine game set. In this manner, the first player can prioritize playing in a live game. If a live game fails to match (e.g., at night or during working hours, there are no live players who have selected the first competitive mode, or there are insufficient live players), a target game is selected from the human-machine game set. Because the robots in this embodiment are more diverse, such a human-machine game is less likely to be discovered by the player as a battle against a robot, thereby satisfying the player's gaming needs and reducing player waiting time.

[0086] As a possible implementation, the game history data includes game level data; accordingly, the target round selection module 32 is further configured to execute: if the game level data indicates that the first player's game level is above a set level threshold, select a real-person round from the game round set as the target round corresponding to the first player, and match the first player with a real-person player corresponding to the real-person round based on the game level data; if the game level data indicates that the first player's game level is not above the set level threshold, select a target round corresponding to the first player from the human-computer rounds in the game round set. By selecting a real-person round as the target round for high-level players, the high-level players' gaming abilities can be more fully utilized, thereby improving their flow experience; by selecting a human-computer round as the target round for low-level players, the robot can provide more appropriate difficulty parameters during the game, maintaining the player's win-loss rate within a reasonable range, thereby improving the flow experience of low-level players.

[0087] As one possible implementation, the game history data includes win / loss data from the most recent N game rounds. Accordingly, the target round selection module 32 is further configured to: determine an expected outcome for the current game round based on the win / loss data from the most recent N game rounds; and select a target round corresponding to the first player from the human-computer rounds in the game round set based on the expected outcome. This method of determining the expected outcome for the current game round based on the win / loss data from the player's most recent N game rounds and selecting the target round based on this outcome can make target round selection more reasonable and maintain the player's win / loss rate within a reasonable range.

[0088] As a possible implementation, the target round selection module 32 is further configured to: if the expected outcome is victory for the first player, select a first type of human-machine round from the human-machine rounds in the game round set as the target round for the first player; wherein the first type of human-machine round offers a higher win rate for the first player than other human-machine rounds; and if the expected outcome is loss for the first player, select a second type of human-machine round from the human-machine rounds in the game round set as the target round for the first player; wherein the second type of human-machine round offers a lower win rate for the first player than other human-machine rounds. This method of selecting the first or second type of human-machine round as the target round based on the first player's win-loss information in the expected outcome can ensure that the first player achieves the expected outcome in the target round.

[0089] As a possible implementation, the game history data also includes the type of the previous game round. Accordingly, the target round selection module 32 is further configured to: if the type of the previous game round was a first-type human-machine round or a second-type human-machine round, select a third-type human-machine round from the human-machine rounds in the game round set as the target round for the first player; wherein the first-type human-machine round provides the first player with a higher win rate than other human-machine rounds, the second-type human-machine round provides the first player with a lower win rate than other human-machine rounds, and the third-type human-machine round provides the first player with a win rate between the first-type human-machine rounds and the second-type human-machine rounds. By determining the target round for the current round based on the type of the previous game round, the types of adjacent rounds can be more diverse, enhancing the player's gaming flow experience.

[0090] As a possible implementation, the target game selection module 32 is further configured to: select the live game from the game game set as the target game for the first player, and match the first player with a live player corresponding to the live game; and in response to the first player waiting for the live game for a second set time without matching a live player, select a fourth type of live game from the live game set as the target game for the first player; wherein the fourth type of live game provides the first player with a random win rate. By promptly providing the first player with the fourth type of live game when the time it takes for the live game to not match a live player reaches the second set time, the first player can be prevented from losing patience and logging off due to continued waiting, thereby enhancing their flow experience.

[0091] Referring to FIG4 , based on the apparatus shown in FIG3 , this embodiment provides another apparatus for processing a battle game. The apparatus further includes: a player type determination module 42 configured to determine whether the first player is a cheating player based on the game history data; a first processing module 44 configured to, if the determination result is yes, select a fifth type of human-machine game from a game game set and start a game corresponding to the fifth type of human-machine game for the first player; wherein the other players in the fifth type of human-machine game are all robots or cheating players; and a second processing module 46 configured to, if the determination result is no, execute the target game selection module 32 to select a target game corresponding to the first player from the game game set based on the game history data. The fifth type of human-machine game is a special human-machine game. By setting up this type of game, cheating players can be isolated from real players, preventing cheating players from interfering with real players and thereby ensuring the flow experience of real players.

[0092] As one possible implementation, the game initiation module 34 is further configured to: determine the robot combat style and number of robots in the target round based on the game skill data of each real player in the target round and the first competitive mode; deploy the robots in the target round for the first player based on the robot combat style and number of robots; and initiate the game corresponding to the target round. By determining the robot combat style and number of robots in the target round based on the game skill data of the real players and the first competitive mode, the robots in the target round can be more closely matched to the game skills of the real players, thereby improving the rationality of the game rhythm.

[0093] As a possible implementation, the parameter adjustment module 36 is further configured to: obtain the first player's performance data in the target round; and adjust the difficulty parameter of the robot in the target round based on the performance data and the type of the target round, so that the first player's win rate in the target round meets the expected win rate. Adjusting the robot's difficulty parameter in this manner can make the robot behave more humanlike throughout the game.

[0094] As a possible implementation, the parameter adjustment module 36 is further configured to: determine an adjustment direction for the difficulty parameter of a target robot in the target round based on the reward information in the performance data and the type of the target round; wherein the target robot is either a friend or an enemy of the first player, and the difficulty parameter includes at least one of the following: reaction speed, accuracy in judging the robot's own or enemy's status, or proficiency in skill release; the adjustment direction includes increasing or decreasing the difficulty parameter; and adjusting the difficulty parameter of the target robot based on the determined adjustment direction. This method of adjusting the robot's difficulty parameter can increase the intensity or volatility of the competition between the two players, thereby enhancing the fun of the game.

[0095] Referring to Figure 5 , based on the device shown in Figure 3 , this embodiment provides another device for processing a battle game. The device further includes a data recording module 52 configured to, in response to the completion of the target round, record the game data of the first player in the target round. This game data is used to initiate the next round for the first player. This data collection method allows for convenient and reliable data, providing a reference for matchmaking in subsequent rounds.

[0096] The processing device for fighting games provided in the embodiments of the present disclosure has the same implementation principle and technical effects as those in the aforementioned method embodiments. For the sake of brief description, for matters not mentioned in the embodiments of the processing device for fighting games, reference may be made to the corresponding contents in the aforementioned method embodiments for processing fighting games.

[0097] The term "and / or" herein simply describes an association relationship between associated objects, indicating that three relationships can exist. For example, "A and / or B" can represent the existence of three situations: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" herein refers to any combination of at least two of any one or more of a plurality of items. For example, "at least one of A, B, and C" can represent any one or more elements selected from the set consisting of A, B, and C.

[0098] An embodiment of the present disclosure also provides an electronic device, as shown in Figure 6, which is a structural diagram of the electronic device, wherein the electronic device includes a processor 61 and a memory 60, and the memory 60 stores computer-executable instructions that can be executed by the processor 61. The processor 61 executes the computer-executable instructions to implement the above-mentioned processing method of the fighting game.

[0099] In the embodiment shown in FIG. 6 , the electronic device further includes a bus 62 and a communication interface 63 , wherein the processor 61 , the communication interface 63 and the memory 60 are connected via the bus 62 .

[0100] The memory 60 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is achieved through at least one communication interface 63 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 62 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 62 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one bidirectional arrow is used in Figure 6, but this does not mean that there is only one bus or one type of bus.

[0101] The processor 61 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method may be completed by hardware integrated logic circuits in the processor 61 or by software instructions. The processor 61 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present disclosure may be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor 61 reads the information in the memory and completes the steps of the method for processing the battle game in the above embodiment in combination with its hardware.

[0102] The embodiments of the present disclosure also provide a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by the processor, the computer-executable instructions prompt the processor to implement the above-mentioned method for processing the battle game. The specific implementation can be found in the aforementioned method embodiment, which will not be repeated here.

[0103] The computer program product of the method, device and electronic device for processing a battle game provided in the embodiments of the present disclosure includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the previous method embodiments. For specific implementation, please refer to the method embodiments and will not be repeated here.

[0104] Unless otherwise specifically stated, the relative steps, numerical expressions and numerical values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0105] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the relevant technology or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling 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 method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0106] In the description of this disclosure, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate the description of this disclosure and simplify the description. They do not indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on this disclosure. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0107] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The scope of protection of the present disclosure is not limited thereto. Although the present disclosure has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed in the present disclosure, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure shall be subject to the scope of protection of the claims.

Claims

1. A processing method for a battle game, the method comprising: Responding to the first player selecting the first competitive mode, selecting a target game for the first player from a set of game sessions; wherein, the set of game sessions includes live-player sessions and multiple different types of human-machine sessions; Responding to the target game being a human-machine session, starting the game corresponding to the target game for the first player; During the process of the game in the target game session, dynamically adjusting the difficulty parameter of the robot in the target game session, and controlling the robot to perform game behaviors according to the adjusted difficulty parameter until the game in the target game session ends; wherein, the difficulty parameter affects the intensity of the game in the target game session and / or the winning or losing result of the first player when the game in the target game session ends.

2. The method according to claim 1, wherein Selecting the target game for the first player from the set of game sessions includes: Obtaining the game historical data of the first player; wherein, the game historical data includes game level data and / or the winning or losing data of the most recent N game sessions; N is a positive integer; Selecting the target game for the first player from the set of game sessions according to the game historical data.

3. The method according to claim 2, wherein, The game historical data includes at least one of the following: the historical winning rate of the first player, the performance data in the previous game session, and the acceptance degree of the first player for human-machine matchups; Selecting the target game for the first player from the set of game sessions according to the game historical data includes: Inputting the game historical data into a pre-trained AI matching model to obtain the human-machine matchup benefit value of the first player; Selecting the target game for the first player from the set of game sessions according to the human-machine matchup benefit value.

4. The method according to claim 2, wherein, Selecting the target game for the first player from the set of game sessions according to the game historical data includes: Matching a live player corresponding to the live-player session for the first player according to the game historical data; Responding to the failure of matching the live-player session within the first set duration, selecting the target game for the first player from the human-machine sessions in the set of game sessions.

5. The method according to claim 2, wherein The game historical data includes game level data; Selecting the target game for the first player from the set of game sessions according to the game historical data includes: If the game level data indicates that the game level of the first player is higher than the set level threshold, selecting a live-player session from the set of game sessions as the target game for the first player, and matching a live player corresponding to the live-player session for the first player according to the game level data; If the game level data indicates that the game level of the first player is not higher than the set level threshold, selecting the target game for the first player from the human-machine sessions in the set of game sessions.

6. The method according to claim 4 or 5, wherein, The game historical data includes the winning or losing data of the most recent N game sessions; Selecting the target game for the first player from the human-machine sessions in the set of game sessions includes: Determining the expected result of this game session according to the winning or losing data of the most recent N game sessions; Selecting the target game for the first player from the human-machine sessions in the set of game sessions according to the expected result.

7. The method according to claim 6, wherein, The selecting the target game for the first player from the human-machine sessions in the set of game sessions according to the expected result includes: If the expected result is the victory of the first player, select the first type of human-machine game from the human-machine games in the game set as the target game corresponding to the first player; wherein, the first type of human-machine game offers a higher win rate for the first player than other human-machine games. If the expected result is the defeat of the first player, select the second type of human-machine game from the human-machine games in the game set as the target game corresponding to the first player; wherein, the second type of human-machine game offers a lower win rate for the first player than other human-machine games.

8. The method according to claim 4 or 5, wherein, The game historical data further includes the type of the previous game; the selecting of the target game corresponding to the first player from the human-machine games in the game set includes: If the type of the previous game is the first type of human-machine game or the second type of human-machine game, select the third type of human-machine game from the human-machine games in the game set as the target game corresponding to the first player; wherein, the first type of human-machine game offers a higher win rate for the first player than other human-machine games, the second type of human-machine game offers a lower win rate for the first player than other human-machine games, and the third type of human-machine game offers a win rate for the first player that is between the first type of human-machine game and the second type of human-machine game.

9. The method according to claim 1, wherein The selecting of the target game corresponding to the first player from the game set includes: Select the real-person game from the game set as the target game corresponding to the first player, and match the real-person player corresponding to the real-person game for the first player; In response to the first player waiting for the second set duration and no real-person player being matched for the real-person game, select the fourth type of human-machine game from the human-machine games in the game set as the target game corresponding to the first player; wherein, the fourth type of human-machine game offers a random win rate for the first player.

10. The method according to claim 2, wherein, The method further includes: Judging whether the first player is a cheating player according to the game historical data; If so, select the fifth type of human-machine game from the game set, and start the game corresponding to the fifth type of human-machine game for the first player; wherein, other players in the fifth type of human-machine game are all robots or cheating players; If not, execute the step of selecting the target game corresponding to the first player from the game set according to the game historical data.

11. The method according to claim 1, wherein, Starting the game corresponding to the target game for the first player includes: Determine the combat styles and the number of robots in the target game according to the game level data of each real-person player in the target game and the first competitive mode; Arrange the robots in the target game for the first player according to the combat styles and the number of robots; Start the game corresponding to the target game.

12. The method according to claim 1, wherein, The dynamically adjusting the difficulty parameters of the robots in the target game includes: Obtain the performance data of the first player in the game of the target game; Adjust the difficulty parameters of the robots in the target game according to the performance data and the type of the target game, so that the win rate of the first player in the game of the target game meets the expected win rate.

13. The method according to claim 12, wherein, Adjusting the difficulty parameters of the robots in the target game according to the performance data and the type of the target game includes: Determine the adjustment direction of the difficulty parameter of the target robot in the target game session according to the reward information in the performance data and the type of the target game session; wherein, the target robot is a friendly or enemy of the first player, and the difficulty parameter includes at least one of the following: reaction speed, accuracy of judging its own or the enemy's state, proficiency of skill release; the adjustment direction includes increasing or decreasing. Adjust the difficulty parameter of the target robot according to the determined adjustment direction.

14. The method according to claim 1, wherein The method further includes: In response to the end of the game of the target game session, record the game data of the first player in the game of the target game session; wherein, the game data is used to start the next game session for the first player.

15. A processing device for a battle game, the device includes: A target game session selection module, configured to execute to select the target game session corresponding to the first player from the set of game sessions in response to the first player selecting the first competitive mode; wherein, the set of game sessions includes live game sessions and multiple different types of man-machine game sessions. A game start module, configured to execute to start the game corresponding to the target game session for the first player in response to the target game session being a man-machine game session. A parameter adjustment module, configured to execute to dynamically adjust the difficulty parameter of the robot in the target game session during the progress of the game of the target game session, and control the robot to perform game behaviors according to the adjusted difficulty parameter until the end of the game of the target game session; wherein, the difficulty parameter affects the intensity of the game of the target game session and / or the win-loss result of the first player at the end of the game of the target game session.

16. An electronic device, including a processor and a memory, the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the method according to any one of claims 1 to 14.

17. A computer-readable storage medium, the computer-readable storage medium stores computer executable instructions, and when the computer executable instructions are called and executed by a processor, the computer executable instructions cause the processor to implement the method according to any one of claims 1 to 14.

Images