Game Interaction Control Method, Device, Storage Medium, and Electronic Device

Abstract

The present disclosure relates to a game interaction control method, apparatus, storage medium, and electronic device, and relates to the field of computer technologies. The method determines the game intention of an agent according to game state information, inputs the game state information, game communication information, and game intention into a communication prediction model to obtain the target communication content corresponding to the agent, and then outputs the target communication content, so that the player character determines the game actions to be executed by the agent based on the target communication content. This can not only improve the activity of the game session when a real player battles with the agent, but also enable the real player to understand the game intention of the agent, so as to cooperate with the agent and improve the quality of the game session.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technologies, and in particular, to a game interaction control method, apparatus, storage medium, and electronic device. Background Art

[0002] In the electronic game industry, applying artificial intelligence (AI) to games has become a development trend. With the iteration of game AI technologies, game AI can serve as both a competitive opponent for human players and a game partner who fights side by side with human players in electronic games, which is of great significance for improving the game experience and player activity.

[0003] However, although current game AI can reach the level of top human players in terms of competitive ability, it lacks communication with human players, resulting in the inability of human players to cooperate with game AI. Summary of the Invention

[0004] This Summary of the Invention section is provided to introduce concepts in a brief form, which will be described in detail in the subsequent Detailed Description section. This Summary of the Invention section is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to be used to limit the scope of the claimed technical solution.

[0005] In a first aspect, the present disclosure provides a game interaction control method, including:

[0006] Obtaining game state information and game communication information of a game session;

[0007] Determining a game intention of an agent according to the game state information, where the game intention represents a battle target of the agent;

[0008] Inputting the game state information, the game communication information, and the game intention into a communication prediction model to obtain target communication content corresponding to the agent;

[0009] Outputting the target communication content, where the target communication content is used to enable a player character in the game session to determine a game action to be executed by the agent based on the target communication content.

[0010] In a second aspect, the present disclosure provides a game interaction control apparatus, including:

[0011] A state determination module configured to obtain game state information and game communication information of a game session;

[0012] An intention determination module, configured to determine the game intention of the intelligent agent according to the game state information, where the game intention represents the battle target of the intelligent agent;

[0013] A communication prediction module, configured to input the game state information, the game communication information, and the game intention into a communication prediction model to obtain the target communication content corresponding to the intelligent agent;

[0014] An output module, configured to output the target communication content, where the target communication content is used to enable the player character in the game session to determine the game action to be executed by the intelligent agent based on the target communication content.

[0015] In a third aspect, the present disclosure provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processing device, the steps of the method described in the first aspect are implemented

[0016] In a fourth aspect, the present disclosure provides an electronic device, including:

[0017] A storage device, on which a computer program is stored;

[0018] A processing device, configured to execute the computer program in the storage device to implement the steps of the method described in the first aspect.

[0019] Based on the above technical solutions, by determining the game intention of the intelligent agent according to the game state information, and inputting the game state information, the game communication information, and the game intention into the communication prediction model, the target communication content corresponding to the intelligent agent is obtained, and then the target communication content is output, so that the player character can determine the game action to be executed by the intelligent agent based on the target communication content. This can not only improve the activity of the game session when a real player battles with the intelligent agent, but also enable the real player to understand the game intention of the intelligent agent, so as to cooperate with the intelligent agent and improve the quality of the game session.

[0020] Other features and advantages of the present disclosure will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Combined with the drawings and referring to the following specific implementation manners, the above and other features, advantages, and aspects of the embodiments of the present disclosure will become more obvious. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic, and the original elements and elements are not necessarily drawn to scale. In the drawings:

[0022] Figure 1 It is a schematic diagram of a scenario of a game interaction control method shown according to an exemplary embodiment.

[0023] Figure 2 It is a schematic diagram of a game interaction control method shown according to an exemplary embodiment.

[0024] Figure 3 It is a schematic structural diagram of an action prediction model shown according to an exemplary embodiment.

[0025] Figure 4a It is a schematic diagram of a game session shown according to an exemplary embodiment.

[0026] Figure 4b It is a schematic diagram of a game session shown according to another exemplary embodiment.

[0027] Figure 5 It is a schematic structural diagram of a communication prediction model shown according to an exemplary embodiment.

[0028] Figure 6 It is a schematic structural diagram of a game interaction control device shown according to an exemplary embodiment.

[0029] Figure 7 It is a schematic structural diagram of an electronic device shown according to an exemplary embodiment. Detailed implementation manners

[0030] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Instead, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0031] It should be understood that the various steps recorded in the method embodiments of the present disclosure can be executed in different orders and / or executed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0032] The term "including" and its variations used herein are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0033] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of the functions executed by these devices, modules or units or the interdependent relationship.

[0034] It should be noted that the modifications of "one" and "multiple" mentioned in this disclosure are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise clearly specified in the context, it should be understood as "one or more".

[0035] The names of the messages or information exchanged between multiple devices in the embodiments of this disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0036] It can be understood that before using the technical solutions disclosed in the embodiments of this disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in this disclosure should be informed to users and the authorization of users should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0037] For example, when a user uses a game AI in a game session, in response to receiving an active request from the user, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information, such as information in the game session. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application program, server, or storage medium that performs the operations of the technical solutions of this disclosure according to the prompt message.

[0038] As an optional but non-limiting implementation manner, the manner of sending a prompt message to the user in response to receiving an active request from the user can be, for example, in the form of a pop-up window. The prompt message can be presented in text in the pop-up window. In addition, the pop-up window can also carry selection controls for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0039] It can be understood that the above process of notifying and obtaining user authorization is only illustrative and does not limit the implementation manners of this disclosure. Other manners that meet relevant laws and regulations can also be applied to the implementation manners of this disclosure.

[0040] At the same time, it can be understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of corresponding laws, regulations and related provisions.

[0041] Before further elaborating on the embodiments of this disclosure, the nouns and terms involved in the embodiments of this disclosure are described. The nouns and terms involved in the embodiments of this disclosure are applicable to the following explanations.

[0042] 1) Game character: It can also be called a virtual object and can be called a hero in some games. It refers to an active object in the game. The active object can be at least one of a virtual character, a virtual animal, and an anime character. Optionally, when the game environment is a three-dimensional game environment, the game character is a three-dimensional solid model. Each game character has its own shape and volume in the three-dimensional game environment and occupies a part of the space in the three-dimensional game environment. Optionally, the game character can be a hero character, a soldier, or a neutral creature in a battle game. In the embodiments of the present disclosure, the game character is taken as an example of a hero character for illustration.

[0043] 2) Agent: It can also be called game AI or man-machine. It refers to a game character that can interact with the game environment in the game. For example, the agent can, in a specific game environment, according to its perception of the game environment, communicate and cooperate or battle with other agents according to existing instructions or through autonomous learning, and autonomously complete the set goals in its game environment. It should be understood that the behavior of the agent in the game is controlled by artificial intelligence applicable to the game and can simulate various behaviors based on the needs of the game.

[0044] The technical solutions of the present disclosure will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] Figure 1 It is a schematic diagram of a scene of a game interaction control method shown according to an exemplary embodiment. As Figure 1 shown, the game interaction control method provided by the embodiments of the present disclosure is applicable to Figure 1 the application scenario shown. The application scenario may include multiple terminals 101 and a server 102. Among them, data interaction can be carried out between the terminal 101 and the server 102 in a wired connection or wireless connection manner.

[0046] The server 102 can be the server 102 of the game platform. The terminal 101 accesses the server 102 through the game client to play the game. For example, the game player can log in to the game through the game client and select an arena mode, arena teammates, arena opponents, etc. to play the game. Among them, the arena teammates and / or arena opponents can be agents, that is, the game session run through the terminal 101 can be a man-machine battle mode or a battle mode with agents participating.

[0047] Taking the human-machine battle mode as an example, in the human-machine battle mode, the terminal 101 can collect the game state information and game communication information of the game session, and send the collected game state information and game communication information to the server 102. The server 102 can receive the game state information and game communication information sent by the terminal 101, and determine the game intention of the intelligent agent according to the game state information. Then, the server 102 inputs the game state information, game communication information and game intention into the communication prediction model to obtain the target communication content corresponding to the intelligent agent. Furthermore, the server 102 sends the target communication content to the terminal 101 so that the terminal 101 controls the intelligent agent to output the target communication content.

[0048] Of course, in the actual application process, the game interaction control method provided by the embodiments of the present disclosure can also be independently executed by the terminal 101. For example, the communication prediction model is deployed in the terminal 101. The terminal 101 obtains the game state information and game communication information of the game session, determines the game intention of the intelligent agent according to the game state information, then inputs the game state information, game communication information and game intention into the communication prediction model to obtain the target communication content corresponding to the intelligent agent, and outputs the target communication content.

[0049] Figure 2 It is a schematic diagram of a game interaction control method shown according to an exemplary embodiment. This method can be executed by Figure 1 the server 102 or the terminal 101 in Figure 2 As shown, this game interaction control method may include the following steps:

[0050] In step 210, the game state information and game communication information of the game session are obtained.

[0051] Here, the game interaction referred to in the embodiments of the present disclosure refers to the communication between the intelligent agent and the real player. The game session can be a session of a battle game or a session of any type with the participation of an intelligent agent. The game session can be a session composed of a real player and an intelligent agent. Among them, the intelligent agent can be a teammate or an opponent of the real player in the game session. Of course, the game session can be a session initially composed entirely of real players, but during the game session, due to the withdrawal of one or more real players, the intelligent agent replaces the real player to control the game character.

[0052] Exemplarily, the game state information includes at least one of the following:

[0053] The role state data of the game character in the game session, the game progress of the game session, and the game environment where the game character is located.

[0054] Among them, the character status data of the game character may include the game camp to which the game character belongs, such as the blue side and the red side. The character status data of the game character may include the health point (HP) of the game character, the mana point (MP) of the game character, the skill status, the economic status, the equipment, and other data. Of course, the character status data may also be represented by other forms of data. For example, the character status data may be the ratio of the remaining health points between game characters, the economic ratio, and so on.

[0055] The game progress of the game session refers to the current time of the game session. For example, the game session progresses to 30 minutes. Of course, the game progress of the game session may also refer to the period of the game session, such as the early stage, the mid-stage, the late stage, the very late stage, etc. The period of the game session can be divided according to the current progress time of the game session. For example, the time from 0 to 10 minutes is divided into the early stage, the time from 10 to 28 minutes is divided into the mid-stage, the time from 28 to 40 minutes is divided into the late stage, and the time after 40 minutes is divided into the very late stage.

[0056] The game environment where the game character is located refers to the terrain, location in the game map, the distance between game characters, and so on.

[0057] It should be noted that the game characters referred to in the embodiments of the present disclosure may include game characters controlled by real players, that is, player characters; the game characters may also include game characters controlled by game AIs, that is, agents.

[0058] It should be understood that the game state information of the game session refers to the characteristic information reflecting the game situation of the game session. The game state information will change over time during the game session. The game state information corresponding to the current moment can be called the current game state information. Generally speaking, the game state information at the current moment is obtained to determine the target communication content of the agent according to the current game situation.

[0059] Exemplarily, the game communication information includes at least one of the following:

[0060] The content of the communication behavior triggered by the game character in the game session, the signal type to which the communication behavior belongs, and the frequency of the communication behavior triggered within a preset time period.

[0061] Among them, the content of the communication behavior triggered by the game character refers to the content of the communication message sent by the game character. The communication behavior can be a communication signal of the marker type, such as an attack signal, a retreat signal, etc. Of course, the communication behavior can also be a voice signal sent in the voice type, a text signal sent in the text type, etc. The signal type to which the communication behavior belongs can be a communication message of types such as the marker type, the voice type, the text type, etc. The frequency of the communication behavior triggered within the preset time period can refer to the frequency of the game character sending communication messages within the preset time period in the game round. Among them, the preset time period can be set according to actual needs, such as the frequency of the game character sending communication messages within the past 100 seconds. It should be noted that the frequency of the game character sending communication messages within the preset time period in the game round can refer to the frequency of the game character sending communication messages of the preset type, or the frequency of sending communication messages of all types.

[0062] It should be understood that the game communication information of the game round refers to the characteristic information used to reflect the communication behavior state between the game characters in the game round. The game communication information in the game round will change over time, and the game communication information corresponding to the current moment can be called the current game communication information. Generally speaking, the game communication information at the current moment is obtained to determine the target communication content of the intelligent agent according to the current communication behavior state.

[0063] In some embodiments, the game state information and the game communication information of the game round can be obtained from the game engine. Among them, the game engine refers to the core components of some pre-written editable computer game systems or some interactive real-time image application programs. Of course, the game state information and the game communication information of the game round can also be obtained by performing image recognition on the game screen of the current frame.

[0064] In step 220, according to the game state information, determine the game intention of the intelligent agent, where the game intention characterizes the battle target of the intelligent agent.

[0065] Here, since the game state information reflects the current game situation of the game round, according to this game state information, the game intention of the intelligent agent in the current game situation can be obtained. The game intention refers to the battle target that the intelligent agent is going to achieve. The battle target can refer to the map area, strategic point or map resource that the intelligent agent is going to defend / attack. For example, the battle target of the intelligent agent can be to help the player character catch people, help the player character clear the minion wave, attack the enemy defense tower, clear the jungle monsters, attack specific monsters in the game (such as BOSS monsters like the Baron Nashor, the Dragon Soul, etc.), obtain the vision of the dark area of the game map, remove the vision ward in the game map, cooperate with the player character to execute a certain game strategy, etc.

[0066] As some examples, game state information can be input into an intention prediction model to obtain game intentions. Among them, the intention prediction model can include an input layer, a hidden layer, and an output layer. This neural network model is based on a policy function and outputs game intentions according to the input game state information. By adjusting the parameters of the hidden layer, the output game intentions can be adjusted.

[0067] As other examples, according to the game camps to which the agents and player characters belong in the game session, the game state information can be adjusted to obtain target state information adapted to the agents, and the target state information is input into the intention prediction model to obtain game intentions.

[0068] Among them, adjusting the game state information according to the game camps to which the agents and player characters belong means modifying the game state information corresponding to each agent to obtain target state information. For example, modifying the HP ratio between the agent and other game characters in the game state information, or modifying the economic difference between the agent and other game characters in the game state information, so as to adjust the game intentions of the agents according to the game situation of the game session.

[0069] For example, for an agent belonging to the opponent's camp, the HP ratio, economic ratio, etc. of the player character relative to this agent can be increased to reduce the offensive desire of the agent in the opponent's camp for the player character, or the HP ratio, economic ratio, etc. of the agent in one's own camp relative to the agent in the opponent's camp can be reduced to increase the offensive desire of the agent in the opponent's camp for the agent in one's own camp.

[0070] Also for example, for an agent in one's own camp, the game intentions of the agent can be modified by adjusting the game state information, so that the agent can cooperate with the player character to achieve the battle goals that the player character is about to achieve, such as cooperating with the player character to clear the minion wave.

[0071] Also for example, for an agent in the opponent's camp, the game intentions of the agent can be modified by adjusting the game state information, so that the agent loses the intention to compete for target strategic resources. For example, when the player character is attacking a resource point, by modifying the game state information of the agent in the opponent's camp, the agent in the opponent's camp is made not to compete with the player character for this resource point.

[0072] As still other examples, the game state information can be used as the input of an action prediction model, and the game intentions of the agent are extracted from the intention prediction layer of the action prediction model.

[0073] Figure 3 It is a schematic structural diagram of an action prediction model shown according to an exemplary embodiment. As Figure 3As shown in the figure, the action prediction model includes a feature extraction layer 301, an intention prediction layer 302, a first fusion layer 303, and an action prediction layer 304. Among them, the feature extraction layer 301 is configured to extract a feature vector from the game state information, the intention prediction layer 302 is configured to obtain the game intention corresponding to the feature vector according to the feature vector, the first fusion layer 303 is configured to fuse the feature vector and the game intention to obtain a first fusion feature, and the action prediction layer 304 is configured to obtain the target operation information corresponding to the agent according to the first fusion feature.

[0074] By extracting the game intention output by the intention prediction layer 302, the game intention corresponding to the agent can be obtained. It should be noted that the intention prediction layer 302 can also be the above-mentioned intention prediction model, and its implemented functions and model logics are the same.

[0075] Among them, the target operation information is used to control the actions of the agent. The target operation information may include the game actions to be executed and the time points for executing the game actions. When the target operation information is to release skill B at time point A, the agent is controlled to release skill B at time point A. When the target operation information is to start moving to target point D along the preset path at time point C, the agent is controlled to start moving along the preset path to target point D at time point C. In step 230, the game state information, the game communication information, and the game intention are input into the communication prediction model to obtain the target communication content corresponding to the agent.

[0076] Here, by inputting the game state information, the game communication information, and the game intention of the agent into the communication prediction model, the target communication content output by the communication prediction model can be obtained. Among them, the target communication content includes the content of the communication message to be sent and the target form of sending the communication message, that is, what kind of communication message is sent in what signal type. For example, the communication message of "attack building XXX" is output in a marked way, the communication message of "help me catch someone" is output in a text way, or the communication message of "request support" is output in a voice form.

[0077] It should be understood that through the game state information, the game communication information, and the game intention, the target communication content that matches the current game situation, the communication behavior state of the game character, and the battle target of the agent can be obtained. The target communication content can reflect the game intention of the agent, so that the real player can understand the game intention of the agent and cooperate with the agent.

[0078] It should be noted that in the embodiments of the present disclosure, the target communication content refers to the communication content corresponding to the agent belonging to the same game camp as the player character. For agents belonging to different game camps from the player character, it may not be necessary to determine the target communication content of such agents. Here, the agent and the player character belonging to the same game camp means that the agent and the player character belong to the same team, and the agent and the player character controlled by the real player are teammates.

[0079] In step 240, output the target communication content, where the target communication content is used to enable the player character in the game session to determine the game actions to be performed by the agent based on the target communication content.

[0080] Here, since the target communication content includes the content of the communication message to be sent and the target form for sending the communication message, the communication message can be output according to the target form. For example, when the target communication content corresponding to the agent is to output the communication message "Request for support" in voice form, then output the communication message "Request for support" in voice form in the user interface. By outputting the target communication content, the player character in the game session can determine the game actions to be performed by the agent based on this target communication content, so that the player character can cooperate with the agent.

[0081] Figure 4a is a schematic diagram of a game session shown according to an exemplary embodiment. As Figure 4a shown, the game characters in the game session include a first agent 401, a second agent 402, a third agent 403, a fourth agent 404, a player character 400, a fifth agent 405, a sixth agent 406, a seventh agent 407, an eighth agent 408, and a ninth agent 409. Among them, the first agent 401, the second agent 402, the third agent 403, the fourth agent 404, and the player character 400 belong to the first game camp, and the fifth agent 405, the sixth agent 406, the seventh agent 407, the eighth agent 408, and the ninth agent 409 belong to the second game camp, and the first game camp and the second game camp are enemy camps to each other.

[0082] When the target communication content of the first agent 401 is to output "Attack resource point 410" in the form of a marking signal, the resource point 410 can be marked by using the signal "Attack 411", so that the player character 400 can understand that the first agent 401 is about to attack the resource point 410 and cooperate with the first agent 401 to attack the resource point 410.

[0083] It should be noted that Figure 4aThe illustrated embodiments are only examples. Which signal type is used to send which communication message in the game depends on the game state information, game communication information, and game intention. For example, when the agent discovers that the game character it is laning against is out of its own vision, the agent can also express in an anthropomorphic tone "The hero XXX has disappeared. Please be careful of being caught."

[0084] Thus, by determining the game intention of the agent according to the game state information, and inputting the game state information, game communication information, and game intention into the communication prediction model to obtain the target communication content corresponding to the agent, and then outputting the target communication content, the player character can determine the game actions that the agent is about to execute based on the target communication content. This can not only improve the activity of the game when real players play against the agent, but also enable real players to understand the game intention of the agent, so as to cooperate with the agent and improve the quality of the game.

[0085] In some embodiments, after determining the target communication content, the agent can be directly controlled to output the target communication content.

[0086] In other embodiments, when the target communication content is associated with the player character, the target communication content is output.

[0087] The target communication content being related to the player character means that the battle goal represented by the target communication content requires the cooperation of the player character. For example, when the player character is in the jungle position, if the target communication content is "The enemy's health is not good. Please assist the jungle to tower dive and catch people", "Our health is not good. Please assist the jungle to clear the minions", etc., it means that the target communication content of the agent is related to the player character.

[0088] Of course, the target communication content being related to the player character can also mean that the player character can cooperate with the agent to achieve its battle goal in the current game state. Among them, the player character can cooperate with the agent to achieve its battle goal in the current game state means that when the agent executes its game intention at a preset position, the player character can reach the preset position and cooperate with the agent to achieve its game intention.

[0089] As Figure 4a shown, when the first agent 401 located at the first position determines the target communication content of "Attack the resource point 410", it can determine that the player character 400 can reach the resource point 410 according to the second position of the player character 400, and the HP state of the player character 400 supports cooperating to attack the resource point 410, then the target communication content is related to the player character 400. At this time, control the first agent 401 to output the target communication content of "Attack the resource point 410", such as marking the resource point 410 by using the signal of "Attack 411".

[0090] It should be noted that whether the target communication content is relevant to the player character mainly depends on whether the game intention corresponding to the target communication content requires the cooperation of the player character to achieve, and / or whether the player character can cooperate with the intelligent agent to achieve its game intention in the current game state. If the target communication content is not relevant to the player character, the target communication content may not be output to avoid outputting too many messages that affect the player character's game experience in the game session.

[0091] In some other embodiments, in the case where multiple intelligent agents in the game session simultaneously generate the corresponding target communication content, at least one target communication content is selected from the multiple target communication contents according to the degree of association between the multiple target communication contents and the player character, and the at least one selected target communication content is output.

[0092] Here, multiple intelligent agents in the game session simultaneously generating the corresponding target communication content means that multiple intelligent agents belonging to the same game camp as the player character in the game session simultaneously generate the target communication content. At this time, at least one target communication content is selected from the multiple target communication contents according to the degree of association between the multiple target communication contents and the player character, and the selected target communication content is output.

[0093] Among them, the target communication content with the largest degree of association with the player character among the multiple target communication contents can be selected as the target communication content to be output. Of course, the target communication content with an association degree greater than the preset threshold can be selected as the target communication content to be output.

[0094] Exemplarily, the magnitude of the degree of association between the target communication content and the player character can be determined by the distance between the intelligent agent corresponding to the target communication content and the player character. The magnitude of the degree of association is negatively correlated with the distance, that is, the closer the distance, the greater the degree of association.

[0095] Figure 4b It is a schematic diagram of a game session shown according to another exemplary embodiment. As Figure 4b shown, at the current moment, the target communication content generated by the first intelligent agent 401 is "Attack resource point 410", and the target communication content generated by the fourth intelligent agent 404 is "Low health, request support". According to the distance between the player character 400 and the first intelligent agent 401 and according to the distance between the player character 400 and the fourth intelligent agent 404, it can be determined that the degree of association between the target communication content generated by the first intelligent agent 401 and the player character 400 is less than the degree of association between the target communication content generated by the fourth intelligent agent 404 and the player character 400. At this time, the communication content of "Low health, request support" can be output.

[0096] Of course, the degree of association between the target communication content and the player character can also be determined according to a preset value corresponding to the game intention corresponding to the target communication content. For example, for each game intention, a corresponding preset value can be set according to the importance between the game intention and the player character. Different game intentions correspond to different preset values.

[0097] Thus, in the case where multiple agents in the game session simultaneously generate corresponding target communication content, by selecting the target communication content with a high degree of association from the multiple target communication contents for output, it is possible to avoid outputting multiple messages simultaneously during the game session, which may affect the game judgment of real players. Moreover, it is also possible to avoid frequent output of communication content, ensuring that real players can cooperate with the agents while also ensuring that real players will not be overly disturbed in the human-machine battle mode. In some implementable embodiments, the communication prediction model includes:

[0098] A multi-layer perceptron configured to encode the game state information and the game communication information to obtain a state feature encoding;

[0099] A second fusion layer configured to receive the game intention output by the intention prediction layer and fuse the game intention and the state feature encoding to obtain a second fusion feature;

[0100] A communication prediction layer configured to determine the target communication content according to the second fusion feature.

[0101] Here, a multi-layer perceptron (MLP) is a feedforward artificial neural network model. By encoding the game state information and the game communication information through the multi-layer perceptron, features related to the game situation and the communication behavior state of the game characters can be extracted, enabling the communication prediction layer to determine the appropriate communication content based on the state feature encoding in the current game situation and communication behavior state. Moreover, by obtaining the game intention output by the intention prediction layer and fusing the game intention with the state feature encoding, it is possible to assist the communication prediction model in obtaining the appropriate target communication content, and this target communication content can represent the game intention of the agent, thus realizing the output of the battle target of the agent in an anthropomorphic tone.

[0102] Figure 5 is a schematic structural diagram of a communication prediction model shown according to an exemplary embodiment. As Figure 5As shown in the figure, the communication prediction model 510 includes a multi-layer perceptron 511, a second fusion layer 512, and a communication prediction layer 513. The action prediction model 520 includes a feature extraction layer 521, an intention prediction layer 522, a first fusion layer 523, and an action prediction layer 524. Among them, the feature extraction layer 521 is configured to extract a feature vector from the game state information, the intention prediction layer 522 is configured to obtain the game intention corresponding to the feature vector according to the feature vector, the first fusion layer 523 is configured to fuse the feature vector and the game intention to obtain a first fusion feature, and the action prediction layer 524 is configured to obtain the target operation information corresponding to the agent according to the first fusion feature.

[0103] The multi-layer perceptron 511 encodes the game state information and the game communication information to obtain a state feature encoding. The second fusion layer 512 receives the game intention output by the intention prediction layer 522, and fuses the game intention and the state feature encoding to obtain a second fusion feature. The communication prediction layer 513 then determines the target communication content according to the second fusion feature.

[0104] Thus, through the above communication prediction model, it is possible to obtain target communication content that can represent the game intention of the agent and is anthropomorphic, improving the flexibility between the agent and real human players. In some implementable embodiments, the communication prediction model is obtained through the following steps:

[0105] Obtain the historical game state information and the historical game communication information of the historical game session within a preset time period;

[0106] According to the historical game state information and the historical game communication information, determine the historical game intention of the player character in the historical game session within the preset time period;

[0107] Mark the historical game intention, the historical game state information, and the historical game communication information with a preset communication content to obtain a first training sample, where the preset communication content is determined according to the historical communication information;

[0108] Based on the first training sample, train a machine learning model to obtain the communication prediction model.

[0109] Here, the historical game session may refer to a game session composed of real human players. For each historical game session, the historical game state information and the historical game communication information within a preset time period are extracted from the historical game session. Among them, the preset time period may be a duration set according to a preset, such as 5 seconds. The historical game state information and the historical game communication information within the preset time period can reflect the game situation and the communication behavior characteristics of the game characters within the preset time period.

[0110] It should be understood that the meanings of the historical game state information and the historical game communication information are the same as those of the above-mentioned game state information and game communication information, and will not be elaborated here.

[0111] Through the historical game state information and the historical game communication information, the historical game intentions of the player roles in the historical game session within a preset time period can be determined. Among them, the meaning of the historical game intention is the same as that of the above-mentioned game intention. Of course, the historical game intention refers to the battle goals of real players within the preset time period.

[0112] Exemplarily, the historical game state information and the historical game communication information can be input into a game intention prediction model to obtain the historical game intention. Among them, the game intention prediction model is obtained by training a machine learning model with second training samples, and the second training samples include historical game state information and historical game communication information marked with game intentions.

[0113] It is worth noting that the game intention prediction model can include an input layer, a hidden layer, and an output layer. This neural network model is based on a policy function and outputs the historical game intention according to the input historical game state information and historical game communication information. By adjusting the parameters of the hidden layer, the output historical game intention can be adjusted. Regarding how to train the machine learning model with the second training samples, reference can be made to existing model training methods, which will not be elaborated here.

[0114] After obtaining the historical game intention, by marking the corresponding preset communication content for the historical game intention, the historical game state information, and the historical game communication information, the first training sample can be obtained. Furthermore, using the first training sample to train the machine learning model, a communication prediction model can be obtained.

[0115] Among them, the preset communication content is determined according to the historical communication information. That is, the preset communication content is determined according to the historical communication information within the preset time period. Exemplarily, a communication content can be selected from the historical communication information within the preset time period as the preset communication content. Of course, it is also possible to perform semantic understanding on the historical communication information within the preset time period to obtain the language behavior purpose represented by the historical communication information, and generate the preset communication content based on the language behavior purpose.

[0116] It is worth noting that by determining the preset communication content from the historical communication information, the communication prediction model can learn the communication methods and language habits of real players, so that the target communication content output by the communication prediction model can be more anthropomorphic.

[0117] Thus, the communication prediction model obtained from the first training sample enables the agent to communicate with real players with a higher degree of anthropomorphism, improving the gaming experience of real players.

[0118] Figure 6 It is a schematic structural diagram of a game interaction control device shown according to an exemplary embodiment. As Figure 6 shown, the present disclosure embodiment provides a game interaction control device, and this device 600 includes:

[0119] A state determination module 601, configured to obtain game state information and game communication information of a game session;

[0120] An intention determination module 602, configured to determine the game intention of the agent according to the game state information, wherein the game intention represents the battle target of the agent;

[0121] A communication prediction module 603, configured to input the game state information, the game communication information, and the game intention into a communication prediction model to obtain the target communication content corresponding to the agent;

[0122] An output module 604, configured to output the target communication content, wherein the target communication content is used to enable the player character in the game session to determine the game action to be executed by the agent based on the target communication content.

[0123] Optionally, the intention determination module 602 is specifically configured to:

[0124] Use the game state information as the input of an action prediction model, and extract the game intention of the agent from the intention prediction layer of the action prediction model;

[0125] wherein, the action prediction model includes a feature extraction layer, the intention prediction layer, a first fusion layer, and an action prediction layer;

[0126] The feature extraction layer is configured to extract a feature vector from the game state information;

[0127] The intention prediction layer is configured to obtain the game intention corresponding to the feature vector according to the feature vector;

[0128] The first fusion layer is configured to fuse the feature vector and the game intention to obtain a first fusion feature;

[0129] The action prediction layer is configured to obtain the target operation information corresponding to the agent according to the first fusion feature, wherein the target operation information is used to control the action of the agent.

[0130] Optionally, the communication prediction model includes:

[0131] A multi-layer perceptron configured to encode the game state information and the game communication information to obtain a state feature encoding;

[0132] A second fusion layer configured to receive the game intention output by the intention prediction layer and fuse the game intention and the state feature encoding to obtain a second fusion feature;

[0133] A communication prediction layer configured to determine the target communication content according to the second fusion feature.

[0134] Optionally, the communication prediction module 603 includes:

[0135] An acquisition unit configured to acquire historical game state information and historical game communication information of a historical game session within a preset time period;

[0136] A determination unit configured to determine the historical game intention of the player role in the historical game session within the preset time period according to the historical game state information and the historical game communication information;

[0137] A marking unit configured to mark the historical game intention, the historical game state information, and the historical game communication information with a preset communication content to obtain a first training sample, where the preset communication content is determined according to the historical communication information;

[0138] A training unit configured to train a machine learning model based on the first training sample to obtain the communication prediction model.

[0139] Optionally, the determination unit is specifically configured to:

[0140] Use the historical game state information and the historical game communication information as the input of a game intention prediction model to obtain the historical game intention;

[0141] where the game intention prediction model is obtained by training a machine learning model with a second training sample, and the second training sample includes historical game state information and historical game communication information marked with game intentions.

[0142] Optionally, the game state information includes at least one of the following:

[0143] The role status data of the game characters in the game session, the game progress of the game session, and the game environment where the game characters are located, the blood volume ratio of the game characters in the game session, the economic difference between the game characters, the levels of the game characters, the game progress of the game session, the position information of the game characters, and the distance between the game characters.

[0144] The game communication information includes at least one of the following:

[0145] The content of the communication behavior triggered by the game characters in the game session, the signal type to which the communication behavior belongs, and the frequency of the communication behavior triggered within a preset time period.

[0146] Optionally, the output module 604 is specifically configured to:

[0147] Output the target communication content when the target communication content is associated with the player character;

[0148] When multiple agents in the game session simultaneously generate corresponding target communication contents, select at least one target communication content from the multiple target communication contents according to the degree of association between the multiple target communication contents and the player character, and output the at least one selected target communication content.

[0149] Regarding the device 600 in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0150] Next, refer to Figure 7 , which shows a schematic structural diagram of an electronic device (such as Figure 1 the terminal or server) 700 suitable for implementing the embodiments of the present disclosure. The terminal in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The electronic device shown is merely an example and should not impose any limitation on the functions and usage scope of the embodiments of the present disclosure.

[0151] As Figure 7As shown, the electronic device 700 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 701, which may perform various appropriate actions and processes according to a program stored in the read-only memory (ROM) 702 or a program loaded from the storage device 708 into the random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 are also stored. The processing device 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704. The input / output (I / O) interface 705 is also connected to the bus 704.

[0152] Generally, the following devices may be connected to the I / O interface 705: an input device 706 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 707 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 708 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 709. The communication device 709 may allow the electronic device 700 to communicate with other devices wirelessly or wirelesly to exchange data. Although Figure 7 an electronic device 700 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. Instead, more or fewer devices may be implemented or had.

[0153] Specifically, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network through the communication device 709, or installed from the storage device 708, or installed from the ROM 702. When the computer program is executed by the processing device 701, the above functions defined in the methods of the embodiments of the present disclosure are executed.

[0154] It should be noted that the computer-readable medium described above in the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0155] In some embodiments, the terminal and the server can communicate using any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0156] The above computer-readable medium can be included in the above electronic device; or it can exist separately without being assembled into the electronic device.

[0157] The above computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: obtain game state information and game communication information of a game session; determine a game intention of an agent according to the game state information, where the game intention represents a battle target of the agent; input the game state information, the game communication information, and the game intention into a communication prediction model to obtain target communication content corresponding to the agent; and output the target communication content, where the target communication content is used to enable a player character in the game session to determine a game action to be executed by the agent based on the target communication content.

[0158] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages or combinations thereof. The programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0159] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that, in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0160] The modules involved in the embodiments of the present disclosure can be implemented in software or in hardware. In some cases, the name of a module does not constitute a limitation on the module itself.

[0161] The functions described above herein can be performed, at least in part, by one or more hardware logic components. By way of example, and without limitation, the types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0162] In the context of the present disclosure, a machine-readable medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0163] The above description is only of the preferred embodiments of the present disclosure and an illustration of the applied technical principles. Those skilled in the art should understand that the scope of the disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by the mutual replacement of the above features with the technical features (but not limited to) having similar functions disclosed in the present disclosure.

[0164] In addition, although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments may also be implemented combinatorially in a single embodiment. Conversely, the various features described in the context of a single embodiment may also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0165] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. On the contrary, the specific features and acts described above are merely example forms for implementing the claims. Regarding the apparatus in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated here.

Claims

1. A game interaction control method, characterized in that, Including: Obtaining game state information and game communication information of a game session; Determining the game intention of an agent according to the game state information, where the game intention represents the battle target of the agent; Inputting the game state information, the game communication information, and the game intention into a communication prediction model to obtain the target communication content corresponding to the agent; Outputting the target communication content, where the target communication content is used to enable the player character in the game session to determine the game action to be executed by the agent based on the target communication content.

2. The method according to claim 1, characterized in that, The determining the game intention of the agent according to the game state information includes: Taking the game state information as the input of an action prediction model, and extracting the game intention of the agent from the intention prediction layer of the action prediction model; Wherein, the action prediction model includes a feature extraction layer, the intention prediction layer, a first fusion layer, and an action prediction layer; The feature extraction layer is configured to extract a feature vector from the game state information; The intention prediction layer is configured to obtain the game intention corresponding to the feature vector according to the feature vector; The first fusion layer is configured to fuse the feature vector and the game intention to obtain a first fusion feature; The action prediction layer is configured to obtain the target operation information corresponding to the agent according to the first fusion feature, where the target operation information is used to control the action of the agent.

3. The method according to claim 2, characterized in that, The communication prediction model includes: A multi-layer perceptron configured to encode the game state information and the game communication information to obtain a state feature encoding; A second fusion layer configured to receive the game intention output by the intention prediction layer, and fuse the game intention and the state feature encoding to obtain a second fusion feature; A communication prediction layer configured to determine the target communication content according to the second fusion feature.

4. The method according to any one of claims 1 to 3, characterized in that, The communication prediction model is obtained through the following steps: Obtaining historical game state information and historical game communication information of a historical game session within a preset time period; Determining the historical game intention of the player character in the historical game session within the preset time period according to the historical game state information and the historical game communication information; Marking the historical game intention, the historical game state information, and the historical game communication information with a preset communication content to obtain a first training sample, where the preset communication content is determined according to the historical communication information; Training a machine learning model based on the first training sample to obtain the communication prediction model.

5. The method according to claim 4, characterized in that, The determining the historical game intention of the player character in the historical game session within the preset time period according to the historical game state information and the historical game communication information includes: Inputting the historical game state information and the historical game communication information into a game intention prediction model to obtain the historical game intention; Among them, the game intention prediction model is obtained by training a machine learning model with a second training sample, and the second training sample includes historical game state information and historical game communication information marked with game intentions.

6. The method according to any one of claims 1 to 3, characterized in that, The game state information includes at least one of the following: The character state data of the game characters in the game session, the game progress of the game session, and the game environment where the game characters are located; The game communication information includes at least one of the following: The content of the communication behavior triggered by the game characters in the game session, the signal type to which the communication behavior belongs, and the frequency of the communication behavior triggered within a preset time period.

7. The method according to claim 1, characterized in that, Outputting the target communication content includes: When the target communication content is associated with the player character, outputting the target communication content; When multiple agents in the game session simultaneously generate corresponding target communication contents, according to the degree of association between the multiple target communication contents and the player character, selecting at least one of the target communication contents from the multiple target communication contents, and outputting the at least one selected target communication content.

8. A game interaction control device, characterized in that, It includes: A state determination module configured to obtain the game state information and game communication information of a game session; An intention determination module configured to determine the game intention of an agent according to the game state information, where the game intention represents the battle target of the agent; A communication prediction module configured to input the game state information, the game communication information, and the game intention into a communication prediction model to obtain the target communication content corresponding to the agent; An output module configured to output the target communication content, where the target communication content is used to enable the player character in the game session to determine the game action to be executed by the agent based on the target communication content.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processing device, it implements the steps of the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, It includes: A storage device on which a computer program is stored; A processing device for executing the computer program in the storage device to implement the steps of the method according to any one of claims 1 to 7.

Images