Non-user role playing character control method and device, server and storage medium

Abstract

The present disclosure provides a non-user role-playing type role control method, device, server and storage medium, which can first detect activity event information triggered by a controlled virtual role (which can be referred to as a player) when the virtual role is active in a virtual environment, generate a structured event package corresponding to a witness NPC based on the current positions of each NPC in the virtual environment and the positional relationship between the occurrence position of the activity event information in the virtual space, simulate the multi-hop propagation of the activity event information in the social relationship graph with the witness NPC as the propagation source of the activity event information based on the above structured event package and the social relationship graph formed by the social relationship of each NPC in the virtual environment, and introduce dynamic distortion at each propagation jump, and then generate the behavior instructions of each NPC based on the individual cognitive memory database of each NPC. The present application can drive each role to form dynamic and differentiated cognition, and ultimately promote the natural emergence of complex interaction behaviors and player associations in the social system.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a control method, device, server, and storage medium for non-user-played roles. Background Technology

[0002] Virtual environments (such as games) in related technologies generally suffer from a lack of social interaction, primarily due to the absence of simulations of information dissemination within social networks and the lack of mechanisms for non-user-played characters (NPCs) to form differentiated perceptions based on incomplete or potentially distorted information. NPCs' "intelligence" is limited to individual reactions to directly observed events, failing to simulate key social phenomena such as "rumor spread" and "group attitude formation" in real society. Furthermore, many related technologies employ a centralized event broadcasting mechanism, where game logic detects an event and uniformly notifies all relevant NPCs. This contradicts the principle in the real world that individuals obtain information based on their location and perception capabilities, resulting in both unrealistic portrayals and a waste of computational resources. Summary of the Invention

[0003] In view of this, in order to solve at least some of the above-mentioned technical problems, this disclosure provides a control method, device, server and storage medium for non-user-played roles.

[0004] According to a first aspect of the present disclosure, a control method for non-user-played roles is provided, wherein the control method is applied in a server of a virtual environment engine, the virtual environment engine being used to provide a virtual environment for at least one client, the control method comprising: The system detects activity event information triggered when a controlled virtual character is active in the virtual environment; wherein, the controlled virtual character refers to a virtual character controlled by the client in the virtual environment. Using each non-user-played role in the virtual environment as an autonomous motion information sensor, a structured event package corresponding to the witness non-user-played role is generated based on the current position of each non-user-played role in the virtual environment and the positional relationship of the occurrence location of the activity event information in the virtual space; wherein, the perception area corresponding to the current position of the witness non-user-played role includes the occurrence location of the activity event information, and the structured event package includes basic event elements and basic emotional influence vector. Based on the social relationship graphs of the non-user-played roles in the virtual environment and the structured event packages, the non-user-played witnesses are used as the propagation source to simulate the multi-hop propagation of the event information in the social relationship graphs. Dynamic distortion is introduced at each propagation jump to update the individual cognitive memory databases of the non-user-played roles in the virtual environment. Based on the individual cognitive memory database corresponding to each of the non-user-played roles in the virtual environment, behavioral instructions for each of the non-user-played roles are generated.

[0005] Optionally, The basic elements of the event include type, subject, object, location, and intensity; the basic emotional influence vector includes the identifier of the non-user-played role of the witness, the environmental snapshot at the time of witnessing, and the perceived quality value; the social relationship graph uses a directed weighted graph G=(V, E, W) to store the social relationships between the non-user-played roles in the virtual space; where node V represents a non-user-played role, including its movement information and individual perception parameters; edge E represents a relationship, and weight W represents the relationship strength; The simulation of the multi-hop propagation of the activity event information in the social relationship graph includes: For each source non-user role whose activity event information is known, iterate through its social connections; For each target non-user role, calculate the propagation probability; wherein, the propagation probability is a function of the relationship and strength between the source non-user role and the target non-user role, the strength of the activity event information, the propagation intention of the source non-user role and the perceived quality of the activity event information, and the current hop count. If the propagation probability is greater than a set probability threshold, the activity event information is transmitted from the source non-user role to the target non-user role.

[0006] Optionally, the introduction of dynamic distortion at each propagation jump includes: Based on the activity event information, propagation hop count, personality tags of the source non-user role, environmental context when the event occurred, and perceived quality value in the structured event package, prompt words are constructed; The prompt words are input into a large language model, which outputs a natural language text that is a distorted description of the environment at the time of witnessing the event, which is consistent with the personality of the non-user role and the event information. The text also includes an additional emotional impact offset due to the description changes and a confidence decay coefficient determined by the degree of exaggeration.

[0007] Optionally, the intensity of the activity event information when calculating the propagation probability each time decays exponentially with the number of propagation hops, and is multiplied by the confidence decay coefficient corresponding to the previous hop count.

[0008] Optionally, the individual cognitive memory database includes: Relationship cognition sub-library: Stores the multi-dimensional emotional attitude vectors of the current non-user role towards other roles. The multi-dimensional emotional attitude vectors are dynamically updated based on the received information content and the personality sensitivity matrix of the current non-user role. Fact Memory Sub-Library: Stores memory records of activity event information known to the current non-user-played role. Each memory record includes the event description text, semantic tags of the description text, source and quality tags, environmental context snapshot, and confidence level. Social Memory Sub-library: Stores references to experiences shared by the current non-user-played role with other roles; The current non-user role is the non-user role corresponding to the individual cognitive memory database.

[0009] Optionally, generating behavioral instructions for each non-user-played role based on the individual cognitive memory database corresponding to each non-user-played role in the virtual environment includes: Obtain the emotional vector of the non-user-played character towards the current interaction target, the memory list related to the current situation, the personality parameters of the non-user-played character, and the current environmental context; Filter candidate intents from a predefined library of behavioral intents; A comprehensive utility is calculated for each candidate intention; wherein the comprehensive utility is obtained by weighted summation of multiple factors: the degree of fit between the candidate intention and the personality of the non-user role, the expected impact of the candidate intention on the emotional relationship between the non-user role and the current interaction target, the strength of the association between the candidate intention and the memory information in the memory list, and the suitability of the candidate intention to the current environment and the current time. The candidate intent with the highest overall utility is determined as the target intent; Based on the target intent and the personality tags of the non-user role, select a target template from the dialogue template library; Based on the memory information corresponding to the target intent and the target template, behavioral instructions for the non-user-played role are generated; wherein, the behavioral instructions include at least one of the following: dialogue text, facial animation, and game action instructions.

[0010] Optionally, the sensing area includes at least one of the following: Planar sector area, 3D spatial sector cone area, circular area, spherical area.

[0011] According to a second aspect of the present disclosure, a control device for non-user-played roles is provided, the control device being applied in a server of a virtual environment engine, the virtual environment engine being used to provide a virtual environment for at least one client, the control device comprising: The event capture and preprocessing module is used to detect activity event information triggered when the controlled virtual character is active in the virtual environment; wherein, the controlled virtual character refers to the virtual character controlled by the client in the virtual environment; The event capture and preprocessing module is further configured to use each non-user-played role in the virtual environment as an autonomous motion information sensor, and generate a structured event package corresponding to the witness non-user-played role based on the current position of each non-user-played role in the virtual environment and the positional relationship of the occurrence location of the activity event information in the virtual space; wherein, the perception area corresponding to the current position of the witness non-user-played role includes the occurrence location of the activity event information, and the structured event package includes basic event elements and basic emotional influence vectors; The propagation and updating module is used to simulate the multi-hop propagation of the activity event information in the social relationship graph based on the social relationship graph of each non-user role in the virtual environment and the structured event package, with the non-user role of the witness as the propagation source, and introduce dynamic distortion at each propagation jump to update the individual cognitive memory database of each non-user role in the virtual environment. The behavior decision-making and content generation module is used to generate behavior instructions for each of the non-user-played roles based on the individual cognitive memory database corresponding to each of the non-user-played roles in the virtual environment.

[0012] According to a third aspect of the present disclosure, a server is provided, the server comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to execute a control method for non-user-played roles as described in any of the first aspects.

[0013] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein when instructions in the storage medium are executed by a processor of a server, the server is enabled to perform a control method for a non-user-played role as described in any of the first aspects.

[0014] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: In this disclosure, activity event information triggered when a controlled virtual character (hereinafter referred to as a player) moves in a virtual environment is detected. Each non-user-played character (NPC) in the virtual environment is used as an autonomous movement information sensor. Based on the current position of each NPC in the virtual environment and the positional relationship between the location of the activity event information and the location in the virtual space, a structured event package corresponding to the witness NPC is generated, thereby achieving distributed perception of the activity event information. Specifically, if the perception area corresponding to the current position of an NPC includes the location of the activity event information, then the corresponding NPC is designated as a witness NPC, and a structured event package corresponding to the target NPC is generated. The structured event package may include a basic emotional influence vector and the basic event elements of the aforementioned activity event information. Then, based on the aforementioned structured event package and the social relationship graph formed by the social relationships of various NPCs in the virtual environment, the witness NPCs serve as the source of information dissemination for the aforementioned events. The multi-hop propagation of event information within the social relationship graph is simulated, and dynamic distortion is introduced at each propagation jump to achieve realistic dissemination of event time information among NPCs in the virtual environment. This updates the individual cognitive memory database of each NPC in the virtual environment. Finally, behavioral instructions for each NPC are generated based on their individual cognitive memory database. This application simulates the limitations, subjectivity, and evolutionary process of information dissemination in real society. NPCs react to events in vastly different ways, and the world presents an organic and dynamic social texture. This invention constructs a computable social network of characters and their relationships in the virtual world and simulates their dynamic discovery of events based on their own spatial perception range by endowing each NPC with independent movement perception capabilities (information sensor). Based on this, by simulating the multi-hop propagation of information, content distortion, and individual memory processes within the network, this application drives each character to form dynamic and differentiated cognitions. Ultimately, this leads to the natural emergence of complex interactive behaviors and organic connections between players within the social system. Based on simple rules and algorithms, this application can automatically generate complex, dynamic, and environmentally intertwined social narratives. Each player experiences a unique history of social relationship development due to differences in the timing of their actions and the location of NPCs. Furthermore, the long-term and unpredictable consequences of player actions make managing personal reputation, influencing community opinion, and understanding and manipulating social networks a core game strategy, greatly enriching the gameplay dimensions.

[0015] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0017] Figure 1 This is a flowchart illustrating a control method for a non-user-played role according to an exemplary embodiment.

[0018] Figure 2 This is a block diagram of a control device for a non-user-played role according to an exemplary embodiment.

[0019] Figure 3 This is a block diagram of a server according to an exemplary embodiment. Detailed Implementation

[0020] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.

[0021] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0022] To facilitate understanding of the embodiments of this application, the following will provide further explanation and description with reference to the accompanying drawings and specific embodiments. These embodiments do not constitute a limitation on the embodiments of this application.

[0023] To address at least some of the aforementioned technical problems, this disclosure provides a control method, apparatus, server, and storage medium for non-user-played roles.

[0024] In this disclosure, activity event information triggered by a controlled virtual character (hereinafter referred to as "player") during activity in a virtual environment is detected. Each non-user-played role (NPC) in the virtual environment is used as an autonomous movement information sensor. Based on the current position of each NPC in the virtual environment and the positional relationship of the activity event information's occurrence location in virtual space, a structured event package corresponding to the witness NPC is generated, thereby achieving distributed perception of the activity event information. Specifically, if the perception area corresponding to the current position of an NPC includes the occurrence location of the activity event information, then the corresponding NPC is designated as a witness NPC, and a structured event package corresponding to the target NPC is generated. The structured event package may include a basic emotional influence vector and the basic elements of the aforementioned activity event information. Then, based on the aforementioned structured event package and the social relationship graph formed by the social relationships of each NPC in the virtual environment, using the witness NPC as the propagation source of the aforementioned activity event information, multi-hop propagation of the activity event information in the social relationship graph is simulated. Dynamic distortion is introduced at each propagation jump, thereby achieving realistic propagation of activity time information among each NPC in the virtual environment, thus updating the individual cognitive memory database of each NPC in the virtual environment. Finally, based on the individual cognitive memory database of each NPC, behavioral instructions for each NPC are generated. This application simulates the limitations, subjectivity, and evolution of information dissemination in real society. NPCs react to events in vastly different ways, presenting the world with an organic and dynamic social texture. This application constructs characters and their relationships in the virtual world into a computable social network, and simulates their dynamic discovery of events based on their spatial perception range by endowing each NPC with independent movement perception capabilities (information sensor). Based on this, by simulating the multi-hop propagation of information, content distortion, and individual memory processes within this network, it drives each character to form dynamic and differentiated cognitions, ultimately prompting the social system to naturally generate complex interactive behaviors and organic connections between players. Based on simple rules and algorithms, this application can automatically generate complex, varied, and environmentally intertwined social narratives. Each player will experience a unique history of social relationship development due to differences in the timing of their actions and the location of NPCs. Furthermore, the consequences of player actions are long-term and unpredictable, making managing personal reputation, influencing community opinion, and understanding and manipulating social relationship networks a core game strategy, greatly enriching the gameplay dimensions.

[0025] In one exemplary embodiment, reference Figure 1 As shown, a control method for non-user-played roles is proposed. This method is applied to a virtual environment engine server, which provides a virtual environment for at least one client. Each client corresponds to one user, and the role controlled by the user in the virtual environment is denoted as the controlled virtual role, or simply "player." The control method includes: S110. Detect activity event information triggered when the controlled virtual character is active in the virtual environment; whereby the controlled virtual character refers to the virtual character controlled by the client in the virtual environment; S120. Using each non-user-played role in the virtual environment as an autonomous mobile information sensor, based on the current position of each non-user-played role in the virtual environment and the positional relationship of the occurrence location of the activity event information in the virtual space, a structured event package corresponding to the non-user-played role of the witness is generated; wherein, the perception area corresponding to the current position of the non-user-played role of the witness includes the occurrence location of the activity event information, and the structured event package includes the basic elements of the event and the basic emotional influence vector. S130. Based on the social relationship graph and structured event package of each non-user role in the virtual environment, with the non-user role of the witness as the propagation source, simulate the multi-hop propagation of event information in the social relationship graph, and introduce dynamic distortion at each propagation jump to update the individual cognitive memory database of each non-user role in the virtual environment. S140. Based on the individual cognitive memory database corresponding to each non-user role in the virtual environment, generate behavioral instructions for each non-user role.

[0026] In step S110, the event capture and preprocessing module can detect player activity events through the game physics / event system and put the detected activity event information (such as collisions) into a global queue for subsequent information propagation.

[0027] In step S120, the event capture and preprocessing module treats each NPC as an autonomous motion information sensor, rather than passively receiving central broadcasts. Each NPC is configured with a fixed movement route or patrol area, as well as individual perception parameters that define the NPC's perception area in the virtual space. The system maintains the real-time position, orientation, and movement status of each NPC.

[0028] The sensing area includes at least one of the following: a planar sector area, a 3D spatial sector-shaped cone area, a circular area, and a spherical area. Of course, the sensing area can also be set as an area of ​​other shapes, and there is no limitation on this.

[0029] The event capture and preprocessing module initiates an independent detection task for each active NPC, which is executed periodically: It calculates the positional relationship between the NPC's current location and the location where the event occurred (i.e., the event's occurrence location) in the virtual space. It then determines whether the occurrence location is within the NPC's configured perception area. This perception area can be designed as different spatial models according to game requirements, such as a planar sector, a 3D spatial sector cone, a circular area, a spherical area, or a combination thereof. If the occurrence location is within the NPC's perception area, it is determined that the NPC witnessed the event. Only for the NPC who witnessed the event, a structured event package with a private perspective is generated.

[0030] The structured event package includes basic event elements and a basic emotional impact vector.

[0031] The basic elements of an event may include type, subject, object, location, intensity, etc.

[0032] The basic emotional impact vector includes the identifier of the witness (not a user-played role), a snapshot of the environment at the time of the witnessing, and a perceived quality value.

[0033] For example, the basic emotional impact vector includes: primary_witness_id: This event packet is generated based on the witnesses of this NPC; environment_context: A snapshot of the environment at the moment of sighting (time, weather, lighting, location type, etc.), provided by the corresponding subsystem of the game engine; perception_quality: A perception quality value (0-1) calculated by considering factors such as spatial distance, environmental factors (e.g., lighting, weather), and line-of-sight obstruction. It is used to quantify the reliability of an eyewitness account. The calculation method for perception quality is related to the specific spatial model of the perception area.

[0034] In step S130, the social relationship graph uses a directed weighted graph G=(V, E, W) to store the social relationships between the non-user-played roles in the virtual space; where node V represents a non-user-played role, including its movement information and individual perception parameters; edge E represents a relationship (such as friends or relatives), and weight W represents the relationship strength, affecting the probability of information propagation. It should be noted that the social relationship graph can be dynamically fine-tuned and persistently stored.

[0035] In the simulation of multi-hop propagation of event information in a social relationship graph, for each source NPC with known event information, its social connections are traversed; for each target non-user role, the propagation probability is calculated; wherein the propagation probability is a function of the relationship and strength between the source NPC and the target NPC, the strength of the event information, the propagation intention of the source NPC, the perceived quality of the event information, and the current number of hops; if the propagation probability is greater than a set probability threshold, the event information is transmitted from the source NPC to the target NPC; if the propagation probability is less than or equal to the set probability threshold, the event information is not transmitted from the source NPC to the target NPC.

[0036] It should be noted that the intensity of the activity event information used in each propagation probability calculation decays exponentially with the number of propagation hops, and is multiplied by the confidence decay coefficient corresponding to the previous hop count. Furthermore, the probability threshold can be set according to actual needs, and its specific value is not limited. For example, if the range of P is 0-1, the probability threshold can be set to 0-0.5, such as 0.2.

[0037] Specifically, when introducing dynamic distortion at each propagation jump, prompt words can be constructed based on the activity event information, propagation jump number, personality tags of the source non-user role, environmental context at the time of the event, and perceived quality value in the structured event package. The prompt words are then input into a set large language model, so that the set large language model outputs a distorted description text in natural language that conforms to the personality of the source non-user role and the environmental atmosphere at the time of witnessing the activity event information, the additional emotional impact offset caused by the description change, and the confidence decay coefficient determined by the degree of exaggeration.

[0038] In some implementations, the propagation and update module can receive structured event packets from different witness NPCs, simulating multi-hop propagation of information in a social relationship graph, and introducing AI-driven dynamic distortion at each propagation jump.

[0039] The propagation simulation process includes: Initialization: Each witness NPC acts as an independent source of information, aware of events from their private perspective; Multi-hop propagation: For each source NPC with known information, traverse its social connections; for each target NPC, calculate the propagation probability P, which is a function of the relationship and strength between the source non-user-played role and the target non-user-played role, the strength of the activity event information, the propagation intention of the source non-user-played role and the perceived quality of the activity event information, and the current hop count.

[0040] If P is greater than the set probability threshold, the information is transmitted from the source NPC to the target NPC, and undergoes distortion processing and intensity attenuation.

[0041] AI-driven dynamic distortion generation: Architecture: Use a lightweight large language model (LLM) module or a service that calls the LLM API.

[0042] Inputs include: source information text, propagation hop count, the narrator NPC's personality tags (e.g., "big mouth," "cautious"), the environmental context of the event, and the perceived quality of the information source. Source information text refers to the text of the event information known to the source NPC. For example, when a witness NPC is the source NPC in this propagation, the source information text is the text of the event information. Similarly, when an intermediate NPC acts as the source NPC during propagation, the source information text is the text of the event information known to that source NPC; in other words, this source information text introduces dynamic distortion from previous propagation processes.

[0043] Processing: Construct a prompt containing all the above inputs, and request the LLM to generate a distorted description in natural language that matches the narrator's personality and the atmosphere of the environment. The "temperature" parameter of the LLM is adjustable to control the level of creativity / exaggeration.

[0044] Output: Distorted descriptive text, offset of additional emotional impact due to changes in description, and confidence decay coefficient determined by the degree of exaggeration.

[0045] Intensity decay: Information intensity decays exponentially with the number of hops, and is multiplied by the confidence decay coefficient caused by distortion.

[0046] For example, given the source event "A knocks down B", in a "rainy night" environment, propagated by an "exaggerated" NPC, the following can be generated by setting a large language model: "My God! In the pitch-black rainy night, A drove like a madman and knocked B flying! Rainwater mixed with blood, it's terrifying!" In addition, in this application, a private, structured individual cognitive memory database can be maintained for each NPC, and the individual cognitive memory database includes: Relationship cognition sub-database: Stores multi-dimensional emotional attitude vectors (such as trust, friendliness, fear) of the current non-user role towards other roles. The multi-dimensional emotional attitude vectors are dynamically updated based on the received information content and the personality sensitivity matrix of the current non-user role. Fact Memory Sub-bank: Stores memory records of activity events known to the current non-user-played role. Each memory record includes the event description text (distorted text), semantic tags for the description text (such as [car accident], [player A], [rainy night], for quick retrieval), source and quality tags (source_type (direct witness / hearsay), primary_witness_id, perception_quality_tags (such as [clear witness], [observed from a distance], [hearsay])), environmental context snapshot, and confidence level; the fact memory sub-bank may also include salience.

[0047] Social Memory Sub-library: Stores references to experiences shared by the current non-user-played role with other roles; The current non-user role is the non-user role corresponding to the individual cognitive memory database.

[0048] In this step, after each propagation jump, the individual cognitive memory database of the target NPC can be updated, thereby ensuring the real-time updating of the individual cognitive memory database of each NPC, thus automatically generating complex, ever-changing social narratives intertwined with the environment.

[0049] In step S140, when an NPC needs to output dialogue text, facial animations, and game action commands, the behavior decision-making and content generation module can generate the NPC's behavior commands based on the individual cognitive memory database corresponding to the NPC, thereby driving the NPC to produce dynamic, differentiated, and organically related behaviors.

[0050] The process involves first acquiring the NPC's emotional vector towards the current interaction target (e.g., the player or other NPCs), a memory list related to the current situation (which can be retrieved through tag matching), the NPC's personality parameters, and the current environmental context. Candidate intentions are then filtered from a predefined behavioral intention library (e.g., friendly greetings, refusing requests, sharing rumors, mentioning third parties). A comprehensive utility U is calculated for each candidate intention, where the comprehensive utility is obtained by weighted summation of multiple factors: the degree of fit between the candidate intention and the personality of the non-user-played character, the expected impact of the candidate intention on the emotional relationship between the non-user-played character and the current interaction target, the correlation strength between the candidate intention and memory information in the memory list, and the suitability of the candidate intention to the current environment and time. Finally, the candidate intention with the highest comprehensive utility is determined as the target intention. Then, based on the target intention and the NPC's personality tags, a target template is selected from a dialogue template library. Based on the memory information corresponding to the target intention and the target template, behavioral instructions for the NPC are generated. These behavioral instructions include at least one of the following: dialogue text, facial animation, and game action instructions.

[0051] When generating NPC behavior commands, a target template can be selected from the dialogue template library based on the target intent and the NPC's personality tags. Then, data (such as third-party character names and event details) is extracted from the NPC's relevant memories and populated into the target template. Furthermore, tone embellishments (such as "I heard...") are automatically added based on the perception_quality_tags in the memories. The bound template is then fed into a style-controllable NLG (Natural Language Generation) component to generate the final conversational dialogue text and output associated facial animations and game action commands. This allows the output of the NPC's action commands.

[0052] Compared with related technologies, this embodiment has the following significant advantages: Creating a highly realistic sense of social immersion: Through distributed perception and distorted propagation, the limitations, subjectivity, and evolution of information dissemination in real society are simulated. NPCs react to events in vastly different ways, presenting the world with an organic and dynamic social texture.

[0053] Achieving low-cost, highly replayable narrative emergence: Based on simple rules and algorithms, the system can automatically generate complex, varied, and environment-interwoven social narratives. Each player will experience a unique history of social relationship development due to differences in the timing of their actions and the location of NPCs.

[0054] This fosters deep social strategy gameplay: the consequences of player actions are long-term and unpredictable. Managing personal reputation, influencing community opinion, and understanding and manipulating social networks become core game strategies, greatly enriching the gameplay dimensions.

[0055] Improve system resource utilization efficiency: Distributed event capture avoids indiscriminate processing of all behaviors. The social simulation engine, as decoupled middleware, uses a relatively simple set of rules and algorithms to sustainably generate almost unlimited social interaction content, resulting in a high return on investment.

[0056] Technological Innovation and Versatility: For the first time in the gaming field, a complete technology chain including "mobile information sensor," "dynamic distortion propagation," and "memory and decision-making based on perception quality" has been systematically proposed. This engine can serve as standard middleware and can be widely applied to various virtual environments requiring high levels of social simulation.

[0057] This invention constructs a three-tiered, general-purpose social simulation engine decoupled from specific game storylines: The bottom layer (perception and relationship layer): Based on a distributed mobile perception architecture, each NPC acts as an independent mobile information sensor, and combines a social relationship graph to describe the persistent connections between NPCs.

[0058] The middle layer (propagation simulation layer) is an information propagation simulation engine built based on communication theory to handle the multi-hop flow of event information in the relationship network and AI-driven dynamic distortion.

[0059] Upper layer (cognitive and behavioral layer): Based on a structured individual cognitive memory database and a behavior decision-making model based on comprehensive utility, it generates NPC behaviors and dialogues based on private cognition.

[0060] In other words, the invention provides a computationally achievable social simulation framework based on social network theory, which can drive non-player characters (NPCs) in games to form organic interactive behaviors based on social information dissemination and individual difference cognition. It is applicable to various virtual environments that require a high degree of social realism, such as role-playing, open world, and social simulation.

[0061] This application constructs a computable social network of characters and their relationships in a virtual world, and simulates the dynamic discovery of events based on their spatial perception range by endowing each NPC with independent movement perception capabilities (information sensor). Based on this, by simulating the multi-hop propagation of information, content distortion, and individual memory processes in this network, it drives each character to form dynamic and differentiated cognition, and ultimately prompts the social system to naturally generate complex interactive behaviors and organic connections between players.

[0062] In one exemplary embodiment, a control method for non-user-played roles is provided. This embodiment is based on a social simulation game.

[0063] System initialization: Define the NPCs and their social relationships in the town. For example: Florist (NPC_C), Baker (NPC_D), Clothing Store Owner (NPC_A), Furniture Store Employee (NPC_B). Set the relationship edges and their weights.

[0064] Configure movement routes and perception parameters for each NPC, which define its perception area in the virtual space (e.g., a fan-shaped area, a circular area, etc.).

[0065] Initialize the individual cognitive memory database for each NPC, with the multidimensional emotion vector set to neutral. The multidimensional emotion vector V = (Friendly, Trusting, Close), with initial values ​​of neutral (0, 0, 0).

[0066] The following is an example of a social simulation of a "car accident on a rainy night".

[0067] Event occurrence and capture: At 22:00 in the game, during heavy rain, Player A hits Player B with their car on the main street.

[0068] NPC_C (Florist): Taking a walk. Based on its configured perception parameters and current location / or orientation, the system determines that the event occurred within its perception area. It is determined to be a witness observation. An event packet E_C is generated, with perception_quality = 0.8 (calculated).

[0069] NPC_D (Baker): Located by the second-floor window, based on its perception parameters (possibly a 3D perception area with height and orientation), the event point is determined to be within its perception area. It is determined to be a witness observation. Event packet E_D is generated, with perception_quality = 0.7 (calculated due to rain and glass).

[0070] NPC_A (Clothing Store Owner): Inside the store, the event point was outside his perception range, and he did not witness it.

[0071] Information dissemination and distortion: The propagation engine uses E_C and E_D as sources to start two propagation threads.

[0072] Thread 1 (NPC_C → NPC_A): Due to NPC_C's high perception quality, the propagation probability is high. Based on NPC_C's "cautious" personality and the rainy night environment, the AI ​​generates a distorted description: "I was just nearby. Player A's car skidded in the rain and hit Player B. It was quite loud." NPC_A receives this information.

[0073] Thread 2 (NPC_D → NPC_B): NPC_D's perception quality is slightly lower. AI generates: "I see from the window that there seems to be a car accident in the alley below. Player A's car hit someone, but the rain is too heavy to see clearly." NPC_B receives this information.

[0074] The information may continue to spread at NPC_A and NPC_B, but the content will become further differentiated.

[0075] Cognitive Update: NPC_A, NPC_B, NPC_C, and NPC_D each create a new record in their respective fact memory sub-banks. The content of each record is a distorted description they received, and they are labeled accordingly (e.g., NPC_A's memory labels include [hearing a paraphrase], [rainy night], and [car accident]).

[0076] Each NPC updates their emotional vector towards Player A and Player B based on the information provided (e.g., NPC_C's trust in Player A decreases more, while NPC_B's decreases less).

[0077] Behavioral Emergence: The next day, Player A visits NPC_A (the clothing store owner): NPC_A retrieves a memory tagged with [Player A] and [car accident], with a high degree of confidence. After calculation by the behavior decision module, the "questioning" intention has high utility. The content generation module selects a questioning template, combines it with memory details, and generates a dialogue: "It was raining so hard last night, how could you drive so carelessly? I heard about it." At the same time, Player A visits NPC_B (furniture shop assistant): NPC_B's retrieved memories are tagged with [fuzzy]. After the decision, the "concerned inquiry" intent has higher utility. The generated dialogue is: "I heard something happened in the alley last night, is it related to you? Is it nothing serious?" Rainy night environment triggers casual conversation: On another rainy night, NPC_C and NPC_D meet on the street. The behavior decision module detects that the current environment highly matches the environmental labels in their memories, and that they have different perspectives on the same event, which is very likely to trigger in-depth communication about the event, and even compare their respective views, driving the evolution of rumors.

[0078] As can be seen from the above embodiments, this invention, starting from simple player behavior, simulates information capture based on individual perception, social network propagation, AI-distorted memory, and differentiated decision-making, ultimately resulting in rich, realistic, and personalized social interactions, greatly enhancing the vitality of the virtual world and the player's immersion. Furthermore, this invention does not limit the specific spatial model of the NPC's perception area, providing flexibility for different game designs and implementations.

[0079] In one exemplary embodiment, a control device for non-user-played roles is provided. This control device is used to implement the control methods of the above embodiments. The control device is applied to a server of a virtual environment engine, which provides a virtual environment for at least one client. (Refer to...) Figure 2 As shown, the control device includes: The event capture and preprocessing module 10 is used to detect activity event information triggered when the controlled virtual character is active in the virtual environment; wherein, the controlled virtual character refers to the virtual character controlled by the client in the virtual environment; The event capture and preprocessing module 10 is further configured to use each non-user-played role in the virtual environment as an autonomous motion information sensor, and generate a structured event package corresponding to the witness non-user-played role based on the current position of each non-user-played role in the virtual environment and the positional relationship of the occurrence location of the activity event information in the virtual space; wherein, the perception area corresponding to the current position of the witness non-user-played role includes the occurrence location of the activity event information, and the structured event package includes basic event elements and basic emotional influence vectors. The propagation and update module 20 is used to simulate the multi-hop propagation of the activity event information in the social relationship graph based on the social relationship graph of each non-user role in the virtual environment and the structured event package, with the non-user role of the witness as the propagation source, and introduce dynamic distortion at each propagation jump to update the individual cognitive memory database of each non-user role in the virtual environment. The behavior decision-making and content generation module 30 is used to generate behavior instructions for each of the non-user-played roles based on the individual cognitive memory database corresponding to each of the non-user-played roles in the virtual environment.

[0080] In one exemplary embodiment, a control device for non-user-played roles is provided. In this embodiment, the basic elements of the event include type, subject, object, location, and intensity; the basic emotional influence vector includes the identifier of the non-user-played role of the witness, an environmental snapshot at the time of witnessing, and a perceived quality value; the social relationship graph uses a directed weighted graph G=(V, E, W) to store the social relationships between the non-user-played roles in the virtual space; wherein, node V represents a non-user-played role, including its movement information and individual perception parameters; edge E represents a relationship, and weight W represents the relationship strength.

[0081] Among them, reference Figure 2 As shown, the propagation and update module 20 includes an information propagation simulation engine 21, which can be used for: For each source non-user role whose activity event information is known, iterate through its social connections; For each target non-user role, calculate the propagation probability; wherein, the propagation probability is a function of the relationship and strength between the source non-user role and the target non-user role, the strength of the activity event information, the propagation intention of the source non-user role and the perceived quality of the activity event information, and the current hop count. If the propagation probability is greater than a set probability threshold, the activity event information is transmitted from the source non-user role to the target non-user role.

[0082] In one exemplary embodiment, a control device for non-user-played roles is provided. This embodiment references... Figure 2 As shown, the information propagation simulation engine 21 can be used for: Based on the activity event information, propagation hop count, personality tags of the source non-user role, environmental context when the event occurred, and perceived quality value in the structured event package, prompt words are constructed; The prompt words are input into a large language model, which outputs a natural language text that is a distorted description of the environment at the time of witnessing the event, which is consistent with the personality of the non-user role and the event information. The text also includes an additional emotional impact offset due to the description changes and a confidence decay coefficient determined by the degree of exaggeration.

[0083] In one exemplary embodiment, a control device that does not involve user-generated roles is provided. In this embodiment, the individual cognitive memory database includes: Relationship cognition sub-library: Stores the multi-dimensional emotional attitude vectors of the current non-user role towards other roles. The multi-dimensional emotional attitude vectors are dynamically updated based on the received information content and the personality sensitivity matrix of the current non-user role. Fact Memory Sub-Library: Stores memory records of activity event information known to the current non-user-played role. Each memory record includes the event description text, semantic tags of the description text, source and quality tags, environmental context snapshot, and confidence level. Social Memory Sub-library: Stores references to experiences shared by the current non-user-played role with other roles; The current non-user role is the non-user role corresponding to the individual cognitive memory database.

[0084] In one exemplary embodiment, a control device for non-user-played roles is provided. In this embodiment, reference is made to... Figure 2 As shown, the behavior decision-making and content generation module 30 can be used for: Obtain the emotional vector of the non-user-played character towards the current interaction target, the memory list related to the current situation, the personality parameters of the non-user-played character, and the current environmental context; Filter candidate intents from a predefined library of behavioral intents; A comprehensive utility is calculated for each candidate intention; wherein the comprehensive utility is obtained by weighted summation of multiple factors: the degree of fit between the candidate intention and the personality of the non-user role, the expected impact of the candidate intention on the emotional relationship between the non-user role and the current interaction target, the strength of the association between the candidate intention and the memory information in the memory list, and the suitability of the candidate intention to the current environment and the current time. The candidate intent with the highest overall utility is determined as the target intent; Based on the target intent and the personality tags of the non-user role, select a target template from the dialogue template library; Based on the memory information corresponding to the target intent and the target template, behavioral instructions for the non-user-played role are generated; wherein, the behavioral instructions include at least one of the following: dialogue text, facial animation, and game action instructions.

[0085] In one exemplary embodiment, a control device for non-user-played roles is provided. In this embodiment, reference is made to... Figure 2 As shown, the propagation and update module 20 may include a social relationship graph management module 22, which can be used to maintain the social relationship graph of each NPC in the virtual space.

[0086] In one exemplary embodiment, a server is provided. The server may have a built-in virtual social simulation engine for simulating a virtual space. The virtual social simulation engine may include an event capture and preprocessing module, a propagation and update module, and a behavior decision and content generation module, etc., for implementing the aforementioned control method.

[0087] Among them, reference Figure 3 As shown, server 100 includes at least one processor 101, memory 102, at least one network interface 104, and user interface 103. The various components in server 100 are coupled together via bus system 105. It is understood that bus system 105 is used to implement communication between these components. In addition to a data bus, bus system 105 also includes a power bus, a control bus, and a status signal bus. However, for clarity, all buses are labeled as bus system 105 in the figure.

[0088] The user interface 103 may include a display, keyboard or click server (e.g., mouse, trackball), touchpad, sensor or touch screen, etc.

[0089] It is understood that the memory 102 in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous DRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The memory 102 described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0090] In some implementations, memory 102 stores elements, executable units or data structures, or subsets thereof, or extended sets thereof: operating system 1021 and application program 1022.

[0091] The operating system 1021 includes various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks. The application program 1022 includes various applications, such as a media player and a browser, used to implement various application functions. Programs implementing the methods of this application embodiment can be included in the application program 1022.

[0092] In this embodiment of the application, the processor 101 executes the method steps provided in each method embodiment by calling the program or instructions stored in the memory 102, specifically the program or instructions stored in the application program 1022.

[0093] The methods disclosed in the embodiments of this application can be applied to or implemented by the processor 101. The processor 101 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware or by instructions in the form of software in the processor 101. The processor 101 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software units in the decoding processor. The software units may be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 102. Processor 101 reads the information in memory 102 and, in conjunction with its hardware, completes the steps of the above method.

[0094] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing servers (DSP devices, DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions of this application, or combinations thereof.

[0095] For software implementation, the techniques described herein can be implemented through units that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented within the processor or external to the processor.

[0096] The server provided in this embodiment can execute all the steps of the above control method, thereby achieving the technical effect of the above control method. For details, please refer to the relevant description of the above control method. For the sake of brevity, it will not be elaborated here.

[0097] This application also provides a storage medium (computer-readable storage medium). This storage medium stores one or more programs. The storage medium may include volatile memory, such as random access memory; it may also include non-volatile memory, such as read-only memory, flash memory, hard disk, or solid-state drive; and it may also include combinations of the above types of memory.

[0098] When one or more programs in the storage medium can be executed by one or more processors to achieve the above-described method of execution on the server side.

[0099] The processor is used to execute programs stored in memory to implement the steps of the methods that are executed on the server side.

[0100] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0101] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0102] It should be noted that the terms "one implementation," "embodiment," "exemplary embodiment," and "some embodiments" used in the specification indicate that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, such phrases do not necessarily refer to the same embodiment. Moreover, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments, whether explicitly described or not, is within the knowledge scope of those skilled in the art.

[0103] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or server that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or server. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or server that includes said element.

[0104] The above embodiments are merely preferred embodiments provided to fully illustrate this application, and the scope of protection of this application is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on this application are all within the scope of protection of this application.

Claims

1. A control method for non-user-played roles, characterized in that, The control method is applied to a server of a virtual environment engine, the virtual environment engine being used to provide a virtual environment for at least one client, and the control method includes: The system detects activity event information triggered when a controlled virtual character is active in the virtual environment; wherein, the controlled virtual character refers to a virtual character controlled by the client in the virtual environment. Using each non-user-played role in the virtual environment as an autonomous motion information sensor, a structured event package corresponding to the witness non-user-played role is generated based on the current position of each non-user-played role in the virtual environment and the positional relationship of the occurrence location of the activity event information in the virtual space; wherein, the perception area corresponding to the current position of the witness non-user-played role includes the occurrence location of the activity event information, and the structured event package includes basic event elements and basic emotional influence vector. Based on the social relationship graphs of the non-user-played roles in the virtual environment and the structured event packages, the non-user-played witnesses are used as the propagation source to simulate the multi-hop propagation of the event information in the social relationship graphs. Dynamic distortion is introduced at each propagation jump to update the individual cognitive memory databases of the non-user-played roles in the virtual environment. Based on the individual cognitive memory database corresponding to each of the non-user-played roles in the virtual environment, behavioral instructions for each of the non-user-played roles are generated.

2. The control method for non-user-played roles according to claim 1, characterized in that, The basic elements of the event include type, subject, object, location, and intensity; the basic emotional influence vector includes the identifier of the non-user-played role of the witness, the environmental snapshot at the time of witnessing, and the perceived quality value; the social relationship graph uses a directed weighted graph G=(V, E, W) to store the social relationships between the non-user-played roles in the virtual space; where node V represents a non-user-played role, including its movement information and individual perception parameters; edge E represents a relationship, and weight W represents the relationship strength; The simulation of the multi-hop propagation of the activity event information in the social relationship graph includes: For each source non-user role whose activity event information is known, iterate through its social connections; For each target non-user role, calculate the propagation probability; wherein, the propagation probability is a function of the relationship and strength between the source non-user role and the target non-user role, the strength of the activity event information, the propagation intention of the source non-user role and the perceived quality of the activity event information, and the current hop count. If the propagation probability is greater than a set probability threshold, the activity event information is transmitted from the source non-user role to the target non-user role.

3. The control method for non-user-played roles according to claim 2, characterized in that, The introduction of dynamic distortion at each propagation jump includes: Based on the activity event information, propagation hop count, personality tags of the source non-user role, environmental context when the event occurred, and perceived quality value in the structured event package, prompt words are constructed; The prompt words are input into a large language model, which outputs a natural language text that is a distorted description of the environment at the time of witnessing the event, which is consistent with the personality of the non-user role and the event information. The text also includes an additional emotional impact offset due to the description changes and a confidence decay coefficient determined by the degree of exaggeration.

4. The control method for non-user-played roles according to claim 3, characterized in that, The intensity of the activity event information when calculating the propagation probability each time decays exponentially with the number of propagation hops, and is multiplied by the confidence decay coefficient corresponding to the previous hop count.

5. The control method for non-user-played roles according to any one of claims 1-4, characterized in that, The individual cognitive memory database includes: Relationship cognition sub-library: Stores the multi-dimensional emotional attitude vectors of the current non-user role towards other roles. The multi-dimensional emotional attitude vectors are dynamically updated based on the received information content and the personality sensitivity matrix of the current non-user role. Fact Memory Sub-Library: Stores memory records of activity event information known to the current non-user-played role. Each memory record includes the event description text, semantic tags of the description text, source and quality tags, environmental context snapshot, and confidence level. Social Memory Sub-library: Stores references to experiences shared by the current non-user-played role with other roles; The current non-user role is the non-user role corresponding to the individual cognitive memory database.

6. The control method for non-user-played roles according to any one of claims 1-4, characterized in that, The generation of behavioral instructions for each non-user-played role based on the individual cognitive memory database corresponding to each non-user-played role in the virtual environment includes: Obtain the emotional vector of the non-user-played character towards the current interaction target, the memory list related to the current situation, the personality parameters of the non-user-played character, and the current environmental context; Filter candidate intents from a predefined library of behavioral intents; A comprehensive utility is calculated for each candidate intention; wherein the comprehensive utility is obtained by weighted summation of multiple factors: the degree of fit between the candidate intention and the personality of the non-user role, the expected impact of the candidate intention on the emotional relationship between the non-user role and the current interaction target, the strength of the association between the candidate intention and the memory information in the memory list, and the suitability of the candidate intention to the current environment and the current time. The candidate intent with the highest overall utility is determined as the target intent; Based on the target intent and the personality tags of the non-user role, select a target template from the dialogue template library; Based on the memory information corresponding to the target intent and the target template, behavioral instructions for the non-user-played role are generated; wherein, the behavioral instructions include at least one of the following: dialogue text, facial animation, and game action instructions.

7. The control method for non-user-played roles according to any one of claims 1-4, characterized in that, The sensing region includes at least one of the following: Planar sector area, 3D spatial sector cone area, circular area, spherical area.

8. A control device for non-user-played roles, characterized in that, The control device is applied in the server of the virtual environment engine, the virtual environment engine being used to provide a virtual environment for at least one client, and the control device includes: The event capture and preprocessing module is used to detect activity event information triggered when the controlled virtual character is active in the virtual environment; wherein, the controlled virtual character refers to the virtual character controlled by the client in the virtual environment; The event capture and preprocessing module is further configured to use each non-user-played role in the virtual environment as an autonomous motion information sensor, and generate a structured event package corresponding to the witness non-user-played role based on the current position of each non-user-played role in the virtual environment and the positional relationship of the occurrence location of the activity event information in the virtual space; wherein, the perception area corresponding to the current position of the witness non-user-played role includes the occurrence location of the activity event information, and the structured event package includes basic event elements and basic emotional influence vectors; The propagation and updating module is used to simulate the multi-hop propagation of the activity event information in the social relationship graph based on the social relationship graph of each non-user role in the virtual environment and the structured event package, with the non-user role of the witness as the propagation source, and introduce dynamic distortion at each propagation jump to update the individual cognitive memory database of each non-user role in the virtual environment. The behavior decision-making and content generation module is used to generate behavior instructions for each of the non-user-played roles based on the individual cognitive memory database corresponding to each of the non-user-played roles in the virtual environment.

9. A server, characterized in that, The server includes: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the control method for non-user-played roles as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the server's processor, the server is able to perform the control method for non-user-played roles as described in any one of claims 1-7.

Images