AI-based NPC control method, device, medium and electronic device

Abstract

The present invention relates to the field of online game technology, and specifically to an AI-based NPC control method, device, medium and electronic device. The AI-based NPC control method includes: when the NPC character is set with first and second type parameters, reading the historical parameters of the NPC character; forming a third parameter based on the historical parameters and the first type parameters; forming a fourth parameter based on the third parameter; adjusting the first type parameters based on the third parameter, and adjusting the second type parameters based on the fourth parameter.

Description

Technical Field

[0001] The present invention relates to the field of online game technology, and in particular to an AI-based NPC control method, device, medium and electronic equipment. Background Art

[0002] In online games, NPC characters are short for non-player characters. They are a type of character in the game that is not controlled by real players. As games become larger and more complex, the performance requirements for NPCs will become higher and higher. The performance of NPCs determines the player's experience of the world in the game. Due to the huge number of NPCs,

[0003] There are many types. If you want to reflect this richness, you need a lot of animation data (such as the implementation method of current 3A games). A large amount of animation data requires a lot of production costs and storage space. In general, the performance of NPC characters ultimately requires processing and modifying the NPC's "daily behavior table", for example: at a certain point in time, where to go, what to do; Time+Position+Action is now mostly a linear daily behavior table for NPC characters, and it is fixed. Unless some specific events are triggered, such as players coming to find NPC characters and triggering the logic of accepting tasks, the basic daily behavior of NPC characters will still not change, resulting in a low sense of richness in the performance of NPC characters and insufficient player experience in the game. Summary of the Invention

[0004] In response to the deficiencies of the existing technology, the present invention provides an AI-based NPC control method, device, medium and electronic device, which can improve the richness of NPC behavior performance based on the influence of player behavior.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0006] In one aspect, the present application provides an AI-based NPC control method, which includes:

[0007] When the NPC character has the first type of parameters and the second type of parameters set, read the historical parameters of the NPC character;

[0008] forming a third parameter based on the historical parameter and the first type of parameters;

[0009] forming a fourth parameter based on the third parameter;

[0010] The first type of parameters are adjusted according to the third parameter, and the second type of parameters are adjusted according to the fourth parameter.

[0011] Preferably, in the above-mentioned AI-based NPC control method, when the NPC character is set with the first and second type parameters, the historical parameters of the NPC character are read, and the historical parameters are all parameters affected by the player's behavior during the player's login time on the NPC character, specifically including:

[0012] Read the player's last login time and behavior parameters, and form a login time difference based on the last login time and the current time;

[0013] Forming a game time that matches the login time difference according to the login time difference;

[0014] Obtaining the influence probability between the behavior parameter and the current NPC character according to the behavior parameter;

[0015] The historical parameter is formed according to the game time difference and the impact probability.

[0016] Preferably, in the above-mentioned AI-based NPC control method, forming the third parameter according to the historical parameter and the first type of parameter specifically includes:

[0017] The first and second impact factors are formed based on historical parameter analysis;

[0018] The third parameter is formed according to the first influencing factor and the second influencing factor.

[0019] Preferably, in the above-mentioned AI-based NPC control method, the first type of parameters includes at least an age coefficient, and the first influencing factor and the second influencing factor formed by analyzing the historical parameters specifically include:

[0020] Obtaining the current age status according to the login time difference and the age coefficient;

[0021] forming a second influencing factor matching the age status according to the age status;

[0022] The first influencing factor is formed by obtaining the influence probability between the behavior parameter and the current NPC character according to the login time difference and the behavior parameter.

[0023] On the other hand, the present application further provides an AI-based NPC control system, which includes:

[0024] A reading unit, for reading historical parameters of the NPC character when the NPC character is set with the first type of parameters and the second type of parameters;

[0025] a third parameter forming unit, forming a third parameter according to the historical parameter and the first type of parameters;

[0026] a fourth parameter forming unit, configured to form a fourth parameter according to the third parameter;

[0027] An adjusting unit adjusts the first type of parameters according to the third parameter and adjusts the second type of parameters according to the fourth parameter.

[0028] Preferably, in the above-mentioned AI-based NPC control system, the historical parameters are all parameters of the impact of the player's behavior on the NPC character during the player's login time, and the reading unit specifically includes:

[0029] A login time difference forming device reads the player's last login time and behavior parameters and forms a login time difference based on the last login time and the current time;

[0030] a game time forming device for forming a game time matching the login time difference according to the login time difference;

[0031] An influence probability forming device, which obtains the influence probability between the behavior parameter and the current NPC character according to the behavior parameter;

[0032] The historical parameter forming device forms the historical parameter according to the game time difference and the impact probability.

[0033] Preferably, in the above-mentioned AI-based NPC control system, the third parameter forming unit specifically includes:

[0034] An analysis device, which analyzes the historical parameters to form a first impact factor and a second impact factor;

[0035] The third parameter forming device forms the third parameter according to the first influencing factor and the second influencing factor.

[0036] Preferably, in the above-mentioned AI-based NPC control system, the first type of parameters at least includes an age coefficient.

[0037] On the other hand, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, it implements any of the above-mentioned AI-based NPC control methods.

[0038] Finally, the present application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-described AI-based NPC control methods when executing the computer program.

[0039] In the above implementation, on the one hand, the actual life time is associated with the game time, and on the other hand, the game time is associated with the status of the NPC character, aiming to improve the authenticity of the NPC character. At the same time, each NPC character changes with the player's behavior, thereby enhancing the fun of the game. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A flow chart of a method for expressing NPC behavior based on the influence of player behavior, provided in Example 1 of the present invention;

[0041] Figure 2 The life cycle curve of an NPC character in a method for expressing NPC behavior based on the influence of player behavior provided in the first embodiment of the present invention;

[0042] Figure 3 A flow chart of a method for expressing NPC behavior based on the influence of player behavior, provided in Example 1 of the present invention;

[0043] Figure 4 A flow chart of a method for expressing NPC behavior based on the influence of player behavior, provided in Example 1 of the present invention;

[0044] Figure 5 This is a structural diagram of an electronic device provided in Example 4 of the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] Example 1

[0047] like Figure 1 As shown, this embodiment provides an AI-based NPC control method, which includes:

[0048] Step S110, when the NPC character is set with the first type of parameters and the second type of parameters, reading the historical parameters of the NPC character;

[0049] The first type of parameters are the basic parameters of the NPC character, for example, the basic parameters may include daily behavior parameters, personality, age, occupation, dreams, willingness to take risks, etc. The second type of parameters are the attitude of the NPC character in interacting with specific players and the task parameter set of the NPC character.

[0050] When a player logs into the game for the first time, both the first and second category parameters of the NPC character are in their initial state. For example, NPC character A's first category parameters are initially set to include daily behavior parameters (eating, sleeping, fishing), an introverted personality, an age of 58, a retired occupation, no dreams, and a very low willingness to take risks. The second category parameters are initially set to include Task A and Task B.

[0051] like Figure 3 As shown, the historical parameters are all parameters of the impact of the player's behavior on the NPC character during the player's login time. The specific formation method includes:

[0052] Step S1101: Read the player's last login time and behavior parameters, and generate a login time difference based on the last login time and the current time;

[0053] Step S1102: A game time period that matches the login time difference is generated based on the login time difference. For example, if the login time difference in real life is 30 days, the game time period that matches the login time difference is 180 days. The matching relationship between the login time difference and the game time period can be adjusted based on actual circumstances. The above is merely an example and does not constitute a specific limitation of this application.

[0054] Step S1103: Obtain the influence probability between the behavior parameter and the current NPC character based on the behavior parameter; the influence probability is related to the current NPC character's living area. A convolutional neural network can be used to calculate this and form an influence probability model.

[0055] For example, if a player secretly kills a pigeon, the probability of this event being spread is relatively low, and the probability of this action affecting all NPC characters is 0.001. If a player kills a pigeon in a square, and there are other NPC characters in the square at the same time, then the player's killing of the pigeon will affect all other NPC characters, but the degree of influence will vary. For example, the probability of affecting animal-loving NPC characters in the square is greater, the probability of affecting the animal-loving NPC characters' friends and family (who did not witness the player's killing of the pigeon in person, but heard about the "player killing the pigeon" event) is moderate, and the probability of affecting NPC characters who do not like animals is smaller. The probability of affecting animal-loving NPC characters is greater than the probability of affecting the friends and family of animal-loving NPC characters.

[0056] Step S1104: forming the historical parameters according to the game time difference and the impact probability.

[0057] Step S120: Form a third parameter based on the historical parameters and the first type of parameters. The first type of parameters at least includes an age coefficient, specifically including:

[0058] Step S1201: Analyze historical parameters to generate a first impact factor and a second impact factor.

[0059] Furthermore, if Figure 4 As shown, step S12011 obtains the current age status based on the login time difference and the age coefficient. Each NPC character is configured with a life cycle curve. Based on the login time difference and the age coefficient, the current NPC character's age status for the current year can be calculated. It should be noted that each NPC character can be configured with an independent life cycle curve, or multiple NPC characters can share a single life cycle curve. This is not a specific limitation.

[0060] Step S12012: forming a second influencing factor that matches the age status according to the age status; referring to the life cycle curve, it can be seen that the perception ability, memory ability, judgment ability and reaction ability of different age statuses are more different.

[0061] Let's list a way to form the second influencing factor. When a player logs into the game for the first time, the first and second parameters of the NPC character are both in their initial state. For example, NPC character A is 58 years old. When the player logs into the game for the second time, one year has passed (perhaps in real life, the time interval between the first and second logins is only two months). At the moment the player logs in, the age of NPC character A has become 59 years old. Specifically, refer to Figure 2 As shown in the life cycle curve of each NPC, when the age of the NPC character changes, its perception parameters, memory parameters, judgment parameters and action parameters will all change.

[0062] Step S12013: Obtain the influence probability between the behavior parameter and the current NPC character according to the time difference and the behavior parameter to form the first influence factor.

[0063] For example, let's consider how the first influencing factor is formed. For example, if a player enjoys fishing in a river, then this behavior will directly or indirectly affect NPC character A's attitude toward the player. Let's set the fishing level coefficient to 1. If the player and NPC character A fish in the same river, and both are within each other's visual range, and if this behavior only occurs once or twice, NPC character A will be largely indifferent to the player. If this behavior occurs daily, NPC character A will be filled with fear of the player, and their trust in NPC character A will be extremely low. This type of influence is called a direct influence. If the player and NPC character A don't fish in the same river, but once during the game they catch a particularly large number of fish, and many people know that the player is a skilled fisherman, NPC character A will develop a certain fear of the player, but their trust in NPC character A will not be directly reduced. This type of influence is called an indirect influence. Because NPC character A is 58 years old and has a certain level of comparison and judgment, the indirect influence will not significantly change the cognition of a 58-year-old adult. Furthermore, the indirect influence also has relatively weak memory. The first impact factor is formed based on direct impact and indirect impact, where the weights of direct impact and indirect impact are different.

[0064] Step S1202: Form the third parameter according to the first influencing factor and the second influencing factor.

[0065] Specifically, an influence coefficient set is formed according to the second influence factor, weight coefficients of direct influence and indirect influence are formed according to the influence coefficient set, and the first influence factor is formed according to the weight coefficient, direct influence and / or indirect influence event level.

[0066] The third parameter is intended to adjust the first type of parameters. As mentioned above, the basic parameters include daily behavior parameters, personality, age, occupation, dreams, and willingness to take risks.

[0067] Daily behavior parameters include eating, sleeping, fishing, and walking. For example, when NPC character A grows from 60 to 65 years old, the number of times he sleeps will gradually decrease. The initial setting for sleeping is 8 hours. As he ages, the sleeping time is reduced to 7 hours. At this time, the extra hour is consumed by other actions in the daily behavior parameters, such as increasing walking time by 20 minutes, eating time by 20 minutes, fishing time by 20 minutes, etc. This allocation method is random. It is also possible to increase walking time by 20 minutes, eating time by 20 minutes, and fishing time by 20 minutes on the first day, and only increase walking time by 10 minutes, eating time by 10 minutes, and fishing time by 40 minutes on the second day. There are no specific restrictions here.

[0068] The sleeping time setting can be set to different times according to different NPC characters and different age groups, and there is no specific restriction here.

[0069] For example, personality also changes with age or the influence of direct or indirect experiences. For example, NPC character B is an animal lover, but he sees or hears about small animals being abused every day, and gradually changes from his original extroverted personality to introverted and even depressed.

[0070] It should be noted that each basic parameter in the first category is set with a parameter value. The personality parameter value, daily behavior parameter value, dream parameter value, and risk-taking parameter value are all directly affected by the third parameter. The personality parameter value, daily behavior parameter value, dream parameter value, and risk-taking parameter value are all related to the current NPC character's experience. The personality parameter value, daily behavior parameter value, dream parameter value, and risk-taking parameter value are not directly related to events.

[0071] Step S130: forming a fourth parameter according to the third parameter;

[0072] The fourth parameter is intended to adjust the second type of parameters, which are the attitude of the NPC character in interacting with a specific player and the task parameter set of the NPC character.

[0073] The third parameter includes a first influencing factor, which is formed based on a weight coefficient and a level of direct and / or indirect influencing events. Direct and / or indirect influencing events are all formed by players, so direct and / or indirect influencing events are all formed by corresponding player information and a first influencing factor formed by player behavior. This first influencing factor directly reacts to the attitude of the NPC character in interacting with a specific player and the task parameter set of the NPC character.

[0074] For example, the first influencing factor can directly influence an NPC's attitude toward a specific player. This attitude includes at least favorability, trust, and fear. For example, if an NPC character particularly trusts the player, they might assign the player a task to babysit children. However, if an NPC character particularly distrusts the player, they might assign the player a task to chop wood.

[0075] The NPC character's task parameter set includes a number of tasks, such as babysitting, that the NPC character trusts the player. Each task type is tagged accordingly, for example, by difficulty. The NPC character assigns tasks of varying difficulty based on their attitude toward the player.

[0076] Step S140: Adjust the first type of parameters according to the third parameter, and adjust the second type of parameters according to the fourth parameter.

[0077] In the above implementation, on the one hand, the actual life time is associated with the game time, and on the other hand, the game time is associated with the status of the NPC character, aiming to improve the authenticity of the NPC character. At the same time, each NPC character changes with the player's behavior, thereby enhancing the fun of the game.

[0078] Example 2

[0079] In another aspect provided by an embodiment of the present invention, the present application further provides an AI-based NPC control system, which includes:

[0080] A reading unit, for reading historical parameters of the NPC character when the NPC character is set with the first type of parameters and the second type of parameters;

[0081] a third parameter forming unit, forming a third parameter according to the historical parameter and the first type of parameters;

[0082] a fourth parameter forming unit, configured to form a fourth parameter according to the third parameter;

[0083] An adjusting unit adjusts the first type of parameters according to the third parameter and adjusts the second type of parameters according to the fourth parameter.

[0084] As a further preferred embodiment, in the above-mentioned AI-based NPC control system, the historical parameters are all parameters of the impact of the player's behavior on the NPC character during the player's login time, and the reading unit specifically includes:

[0085] A login time difference forming device reads the player's last login time and behavior parameters and forms a login time difference based on the last login time and the current time;

[0086] a game time forming device for forming a game time matching the login time difference according to the login time difference;

[0087] An influence probability forming device, which obtains the influence probability between the behavior parameter and the current NPC character according to the behavior parameter;

[0088] The historical parameter forming device forms the historical parameter according to the game time difference and the impact probability.

[0089] Preferably, in the above-mentioned AI-based NPC control system, the third parameter forming unit specifically includes:

[0090] An analysis device, which analyzes the historical parameters to form a first impact factor and a second impact factor;

[0091] The third parameter forming device forms the third parameter according to the first influencing factor and the second influencing factor.

[0092] Preferably, in the above-mentioned AI-based NPC control system, the first type of parameters at least includes an age coefficient.

[0093] The working principle of this embodiment is the same as that of the first embodiment, so it will not be described in detail.

[0094] Example 3

[0095] The present application also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform:

[0096] When the NPC character has the first type of parameters and the second type of parameters set, read the historical parameters of the NPC character;

[0097] forming a third parameter based on the historical parameter and the first type of parameters;

[0098] forming a fourth parameter based on the third parameter;

[0099] The first type of parameters are adjusted according to the third parameter, and the second type of parameters are adjusted according to the fourth parameter.

[0100] Storage medium - any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media, such as CD-ROMs, floppy disks, or tape drives; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (such as hard disks or optical storage); registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the computer system in which the program is executed, or may be located in a different second computer system that is connected to the computer system via a network (such as the Internet). The second computer system may provide program instructions to the computer for execution. The term "storage medium" may include two or more storage media that may reside in different locations (e.g., in different computer systems connected via a network). The storage medium may store program instructions (e.g., embodied as a computer program) that can be executed by one or more processors.

[0101] Of course, the storage medium containing computer-executable instructions provided in an embodiment of the present application is not limited to the NPC behavior performance method based on the influence of player behavior as described above, and can also execute related operations in the NPC behavior performance method based on the influence of player behavior provided in any embodiment of the present application.

[0102] Example 4

[0103] An embodiment of the present application provides an electronic device, into which the NPC behavior representation device based on the influence of player behavior provided by the embodiment of the present application can be integrated. Figure 5 This is a schematic diagram of the structure of an electronic device provided in Example 4 of this application. Figure 5 As shown, this embodiment provides an electronic device 400, which includes: one or more processors 420; a storage device 410 for storing one or more programs. When the one or more programs are executed by the one or more processors 420, the one or more processors 420 implement:

[0104] When the NPC character has the first type of parameters and the second type of parameters set, read the historical parameters of the NPC character;

[0105] forming a third parameter based on the historical parameter and the first type of parameters;

[0106] forming a fourth parameter based on the third parameter;

[0107] The first type of parameters are adjusted according to the third parameter, and the second type of parameters are adjusted according to the fourth parameter.

[0108] like Figure 5 As shown, the electronic device 400 includes a processor 420, a storage device 410, an input device 430, and an output device 440; the number of processors 420 in the electronic device can be one or more. Figure 2 In the figure, a processor 420 is used as an example; the processor 420, the storage device 410, the input device 430 and the output device 440 in the electronic device can be connected via a bus or other means. Figure 2 The connection via bus 450 is taken as an example.

[0109] The storage device 410 is a computer-readable storage medium that can be used to store software programs, computer-executable programs, and module units, such as program instructions corresponding to the NPC behavior performance method based on player behavior influence in the embodiment of the present application.

[0110] The storage device 410 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function; the data storage area may store data created based on the use of the terminal, etc. Furthermore, the storage device 410 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state memory device. In some instances, the storage device 410 may further include a memory remotely located relative to the processor 420, and such remote memory may be connected via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0111] The input device 430 may be used to receive input numbers, character information or voice information, and generate key signal input related to user settings and function control of the electronic device. The output device 440 may include a display screen, a speaker and other devices.

[0112] It should be noted that, in this article, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements that are inherent to such process, method, article or apparatus.

[0113] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An AI-based NPC control method, characterized in that: include: When the NPC character is set with first and second type parameters, the historical parameters of the NPC character are read; the first type parameters are the basic parameters of the NPC character, and the second type parameters are the task parameter set of the NPC character's interaction with a specific player; the historical parameters are all parameters affected by the player's behavior during the player's login time. forming a third parameter according to the historical parameter and the first type of parameters, wherein the third parameter is used to adjust the first type of parameters; forming a fourth parameter according to the third parameter, wherein the fourth parameter is used to adjust the second type of parameters; Among them, when the NPC character is set with the first type of parameters and the second type of parameters, the historical parameters of the NPC character are read, specifically including: reading the behavioral parameters of the player's last login time and login status, forming a login time difference according to the last login time and the current time; forming a game time matching the login time difference according to the login time difference; obtaining the influence probability between the behavioral parameter and the current NPC character according to the behavioral parameter; and forming the historical parameter according to the game time difference and the influence probability.

2. The AI-based NPC control method according to claim 1, characterized in that: Forming the third parameter according to the historical parameter and the first type of parameters specifically includes: The first and second impact factors are formed based on historical parameter analysis; The third parameter is formed according to the first influencing factor and the second influencing factor.

3. The AI-based NPC control method according to claim 2, characterized in that: The first type of parameters includes at least an age coefficient, and the first influencing factor and the second influencing factor formed by analyzing the historical parameters specifically include: Obtaining the current age status according to the login time difference and the age coefficient; forming a second influencing factor matching the age status according to the age status; The first influencing factor is formed by obtaining the influence probability between the behavior parameter and the current NPC character according to the login time difference and the behavior parameter.

4. The AI-based NPC control method according to claim 2, characterized in that: The method for determining the first impact factor includes: forming an impact coefficient set according to the second impact factor; Forming weight coefficients of direct impact and indirect impact based on the impact coefficient set; The first impact factor is formed according to a weight coefficient, a direct impact event level and / or an indirect impact event level.

5. An AI-based NPC control system, characterized in that: include: A reading unit, when the NPC character is set with first-category parameters and second-category parameters, reads historical parameters of the NPC character; the first-category parameters are basic parameters of the NPC character, the second-category parameters are a set of task parameters for the NPC character's interaction with a specific player; the historical parameters are all parameters affected by the player's behavior during the player's login time on the NPC character; a third parameter forming unit, forming a third parameter according to the historical parameter and the first type of parameters; a fourth parameter forming unit, configured to form a fourth parameter according to the third parameter; an adjusting unit, configured to adjust the first type of parameters according to the third parameter and adjust the second type of parameters according to the fourth parameter; Among them, the reading unit specifically includes: a login time difference forming device, which reads the player's last login time and behavior parameters, and forms a login time difference based on the last login time and current time; a game time forming device, which forms a game time matching the login time difference based on the login time difference; an influence probability forming device, which obtains the influence probability between the behavior parameter and the current NPC character based on the behavior parameter; a history parameter forming device, which forms the history parameter based on the game time difference and the influence probability.

6. The AI-based NPC control system according to claim 5, characterized in that: The third parameter forming unit specifically includes: An analysis device, which analyzes the historical parameters to form a first impact factor and a second impact factor; The third parameter forming device forms the third parameter according to the first influencing factor and the second influencing factor.

7. The AI-based NPC control system according to claim 6, characterized in that: The first type of parameters includes at least an age coefficient; The analysis device is specifically used to: obtain the current age status based on the login time difference and the age coefficient; form a second influencing factor matching the age status based on the age status; and obtain the influence probability between the behavior parameter and the current NPC character based on the login time difference and the behavior parameter to form the first influencing factor.

8. The AI-based NPC control system according to claim 6, characterized in that: The analyzing device is specifically used for: forming an impact coefficient set according to the second impact factor; Forming weight coefficients of direct impact and indirect impact based on the impact coefficient set; The first impact factor is formed according to a weight coefficient, a direct impact event level and / or an indirect impact event level.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, an AI-based NPC control method according to any one of claims 1 to 4 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the AI-based NPC control method according to any one of claims 1 to 4 is implemented.

Images