USRE45132E1

Movatterモバイル変換

Info

Publication number: USRE45132E1
Application number: US13/650,071
Authority: US
Inventors: John L. Caporale; Michael J. Caporale
Original assignee: PEA TREE FOUNDATION LLC
Current assignee: Xenogenic Development LLC
Priority date: 2007-03-20
Filing date: 2012-10-11
Publication date: 2014-09-09
Also published as: US20080235581A1; US7814041B2

Abstract

A system operative to execute and train at least one avatar for each user of an interactive environment comprising a knowledge engine operative to continuously monitor each user's response to events in the interactive environment while each user controls the at least one avatar, a knowledge base operative to store each of the monitored user responses to events in the interactive environment and an action engine operative to control one or more actions of the at least one avatar for each user in the interactive environment based on the stored monitored responses regardless of each user's control of the at least one avatar.

Description

FIELD

The present disclosure relates generally to information processing, and in particular but not exclusively, relates to a system and method for control, training and autonomous execution of one or more avatars in an interactive environment.

BACKGROUND

In recent years, the development of new and powerful computing platforms has enabled more powerful computer simulations and computer games to be developed which enable users to engage a wide variety of simulated activities. These computing platforms have also begun to incorporate various forms of artificial intelligence that can enhance a user's experience in an interactive computer environment. Among the range of applications provided by advanced computing platforms are those that include the use of one or more automated personas and intelligent agents. These personas or agents have come to be referred to in certain instances as “avatars” because they may be portrayed in interactive environments with various human-like features.

Research directed to the control and autonomous of operation of intelligent agents in an interactive environment has produced mixed results. A significant degree of research work has been performed to thoroughly understand the mechanics of human motion that can be rendered in a computer generated environment, but such research has provided little assistance in defining the features, methods or techniques to be used to enable these intelligent persona to effectively operate autonomously within an interactive environment regardless of a human user's direct control of the personas. Indeed, research into the most pioneering forms of artificial reasoning is still very much in a nascent stage and little is understood about how to effectively capture, analyze, build associative networks based on a user's past and present behaviors, and apply stored knowledge that is organized in the form of content and related meta-data associative networks to enable the rapid organization, searching and application of such knowledge to the real-time autonomous operation of intelligent personas in an interactive environment.

Thus, there is a current and pressing need for systems and methods that can provide continuous monitoring of events, actively monitor user responses to those events, and apply controlled actions to intelligent persona in a variety of interactive environments regardless of a user's direct involvement in controlling such persona in these environments.

BRIEF DESCRIPTION OF THE DRAWINGS

Non-limiting and non-exhaustive embodiments are described with reference to the following figures, wherein like reference numerals refer to like parts throughout the various views unless otherwise specified.

FIG. 1 is a block diagram illustrating network connectivity between servers and a various client devices in an embodiment.

FIG. 2 is a block diagram illustrating a standalone client device in an embodiment.

FIG. 3 is a block diagram illustrating the components of an action engine in an embodiment.

FIG. 4 is a block diagram illustrating the components of a knowledge engine in an embodiment.

FIG. 5A is a block diagram illustrating the components of a server device in an embodiment.

FIG. 5B is a block diagram illustrating the components of a client device in an embodiment.

FIG. 6A is a block diagram illustrating the components of a server device in an embodiment.

FIG. 6B is a block diagram illustrating the components of a client device in an embodiment.

FIG. 7 is a flow diagram illustrating a method for monitoring, categorizing and storing user responses in an embodiment.

FIG. 8 is a flow diagram illustrating a method for evaluating user responses and controlling the actions of an avatar based on evaluated user responses in an embodiment.

FIG. 9 is a flow diagram illustrating a method for applying pattern analysis to user responses in an embodiment.

FIG. 10 is a flow diagram illustrating a method for alerting a user and monitoring user responses in an embodiment.

FIG. 11 is a flow diagram illustrating a method for controlling the operational mode of an interactive environment in an embodiment.

FIG. 12 is a flow diagram illustrating a method for autonomous operation of avatars in an interactive environment in an embodiment.

FIG. 13 is a flow diagram illustrating a method for determining actions for execution of avatars in a competitive operational mode of an interactive environment in an embodiment.

FIG. 14 is a flow diagram illustrating a method for determining actions for execution of avatars in an assisted operational mode of an interactive environment in an embodiment.

DETAILED DESCRIPTION

In the description to follow, various aspects of embodiments will be described, and specific configurations will be set forth. These embodiments, however, may be practiced with only some or all aspects, and/or without some or these specific details. In other instances, well-known features are omitted or simplified in order not to obscure important aspects of the embodiments.

Various operations will be described as multiple discrete steps in turn, in a manner that is most helpful in understanding each disclosed embodiment; however, the order of description should not be construed as to imply that these operations are necessarily order dependent. In particular, these operations need not be performed in the order of presentation.

The description repeatedly uses the phrases “in one embodiment”, which ordinarily does not refer to the same embodiment, although it may. The terms “comprising”, “including”, “having”, and the like, as used in the present disclosure are synonymous.

FIG. 1 illustrates asystem100 comprised of a plurality ofclient devices104a-104d, anetwork106 and a plurality ofservers102a-102d.Various client devices104a-104d may be included in the system for interactive computing and communications between and among theserver devices102a-102d.Client device104a represents a personal computer,client device104b represents a video gaming station, such as a Microsoft Xbox, Microsoft Xbox 360 or Nintendo play station.Client device104c represents a handheld device such as a personal digital assistant or any of a variety of handheld gaming devices, such as a Sony PlayStation or Nintendo Game Boy, andclient device104d represents a portable computer. Network106 included in thesystem100 may be any of a number of networks including a local area network, a wide area network, the Internet, a wireless network or a combination of networking systems including wired, wireless and optical communication channels and devices.

FIG. 2 illustrates aclient device104 in a standalone embodiment. As shown,client device104 includes one ormore input devices202, acommunication interface204, a ReadOnly Memory206, astorage device208, acentral processing unit214, one ormore output devices210 and aprogram memory212. The one ormore output devices210 can be desktop computer monitors, the displays of laptop computers, the displays of mobile and handheld devices, or the cathode ray tube displays of televisions that are coupled to computer gaming devices (e.g., Xbox, Xbox 360, etc.). Each of these modules is communicatively coupled tocommunication bus216 for inter-module communication.Program memory212 includes several program components, which components comprise a software-implemented system to be used for monitoring user responses and events in an interactive environment and for controlling the actions of one or more avatars in the interactive environment. As shown in this figure, theprogram memory212 includes aknowledge base218, aknowledge engine220, anaction engine222, an interactive computer-generatedenvironment224, and anoperating system226.Central processor214 interacts withprogram memory212 by sending commands and receiving data fromprogram memory212 overcommunication bus216.

In a stand-alone embodiment such as the one depicted inFIG. 2, a game station such as a Sony PlayStation® or Nintendo GameBoy® (client device104b) includesknowledge engine220 to continuously monitor events in theinteractive environment224 and to monitor the user responses to those events in theinteractive environment224.Knowledge engine220 actively analyses, associates and categorizes the user responses and stores those categorized associations among user responses inknowledge base218. In the software-implemented system, one or more avatars may be controlled by a human user in theinteractive environment224. However, theknowledge engine220 actively monitors each user's response in order to build a knowledge base of associations between responses and events which represent the range of user responses to the myriad events occurring in theinteractive environment224. The actual control of an avatar's action in theinteractive environment224 is performed byaction engine222 upon issuance of requests byknowledge engine220.

FIG. 3 illustrates several components comprisingaction engine222. These components are common to each of the alternative embodiments of the software-implemented system. In the specific instance of the standalone alone, as shown,subscriber registration component302 is used for registering each new subscriber or user to enable a registered user to activate and use one or more avatars in theinteractive environment224.Avatar activation component304 is used to activate each new avatar to be used in the in theinteractive environment224 by a registered user. Only registered users are permitted to activate avatars for operations in theinteractive environment224.Action execution component306 is responsible for controlling the actions of avatars in theinteractive environment224 based on one or more requests fromknowledge engine220. The actions performed byaction execution component306 are based on one or more control rules stored in avatarcontrol rules database308.Database308 represents a repository of rules for controlling the actions of avatars in theinteractive environment224.Knowledge engine interface310 is the communication interface betweenknowledge engine220 and theaction engine222 and it is used to receive requests fromknowledge engine220 for execution byaction execution component306. These requests are implemented by one or more control rules stored in the avatarcontrol rules database308.

A particularly important aspect of the software-implemented system is that the action engine may be implemented as an artificial intelligence engine relying upon one or more expert systems, neural networks or other forms of computational intelligence. However, such artificial intelligence engines are necessarily directed by the issuance of requests by theknowledge engine220.Knowledge engine220 is integrated with and actively searches the repository of stored associations between events and user responses to determine what actions should be implemented by an avatar when operating in an autonomous operational mode in theinteractive environment224. Requests are issued by theknowledge engine220 to ensure that theaction execution component306 in theaction engine222 implements actions based on requests which are ultimately based upon previously determined patterns of user responses to events in theinteractive environment224. In this manner, the autonomous actions of one or more of a user's avatars in theinteractive environment224 will most accurately mimic the user's responses with the user is not actively controlling the one or more avatars in theinteractive environment224.

FIG. 4 represents the components provided inknowledge engine220. Although reference is made to theknowledge engine220 in the standalone embodiment, each such component may be included along with all of the other components, or in various operational combinations including fewer of the specified components. In the present embodiment,knowledge engine220 includesevent monitoring component402,response monitoring component404, avatarmode management component406,pattern analysis component408,categorization component410, knowledge basedcontroller412 andaction engine controller414.Event monitoring component402 continuously and actively monitors events occurring in theinteractive environment224. More specifically,event monitoring component402 monitors events in theinteractive environment224 and interoperates withresponse monitoring component404 to enablepattern analysis component408 to form associations between events and user responses as well as categories of user responses according to one or more metrics. These associations among events are stored inknowledge base218 for later searching and a generation of action requests.Response monitoring component404 continuously monitors user responses to the monitored events in theinteractive environment224.

The following is a representative listing of metrics that can be applied bycategorization component410 to the monitored user responses. This list is not exhaustive and is provided herein solely for the purpose of illustrating a representative sample of metrics for application by a knowledge engine in the various embodiments discussed herein. In alternative embodiments, subsets of the listed metrics may be used for user response categorizations, while in other embodiments different or additional metrics may be applied.


	PERFOR-		EMO-
USER RESPONSE	MANCE	DECISION	TION

jump over object		X
go around object		X
use stealth		X
choice of weapons		X
use of tactical options		X
help a team member		X	X
disregard for personal safety		X	X
rude behavior unrelated to game play			X
navigation skill		X
ability to understand choices		X	X
ability to perceive best route to		X	X
objective
marksmanship	X
steering	X
completing a jump	X
time to make a decision	X
time to push necessary button	X
time to respond to stimuli	X
sportsmanship		X	X
moral decision making		X	X
height of jump or leap	X
average number of successful jumps	X
average number of kills	X
use of speed	X
driving skills	X
punching skills	X
swordsmanship	X
learn rate	X
unusual choices against known data		X	X
unusual choices against known		X	X
stimuli
sacrifice to further team objectives		X	X

Interactive environment

224 is a computer-generated environment for use in executing computer simulations or computer video games in which various computer-generated events, or events created by other users active in theinteractive environment224, may occur. These events may trigger or require responses from one or more of a user's avatars when these avatars are operating in an autonomous mode, an assisted duality mode or a competitive duality mode. Avatars can be controlled by users of theclient devices104a-104d or can be executing autonomously or independently of the log-in status of a registered user.

Avatarmode management component406 regulates the control of the avatars based on the operational mode of theinteractive environment224. Each registered user can have one or more avatars for execution in theinteractive environment224; however, the operation of these avatars can vary depending on the operational mode of theinteractive environment224. When a user is not actively logged into theinteractive environment224, the registered user's avatar(s) will still be capable of operating in theinteractive environment224 and responding to events monitored byevent monitoring component402 in an autonomous mode. In a real user mode, each avatar is directly controlled by a registered user whose responses to events in theinteractive environment224 will be monitored byresponse monitoring component404. User responses are the basis of actions performed by avatars in an autonomous mode. A stored profile or history of user response included inknowledge base218 are evaluated byknowledge engine220 when requests are generated and issued toaction engine222 to have specific actions implemented and executed by avatars under the control of theaction engine222.

Pattern analysis component

408 applies pattern analysis techniques based in part on methods that determine the frequency of occurrence of certain user responses to monitored events. These techniques also apply certain statistical relevance ranking and correlations between and among user responses to determine patterns in user responses to monitored events.Pattern analysis component408 forms associations between monitored events and one or more user responses to the monitored event. These associations are determined in real-time and are updated in a dynamic manner based on information compiled byresponse monitoring component404.

After association of user responses to events, the user responses will be categorized according to one or more metrics bycategorization component410. Among the specific metrics used for categorizing user responses are performance metrics, decision metrics and emotion metrics. Emotion metrics are based in part on the performance metrics and the decision metrics. Among the various types of performance metrics are timing measures for each user response monitored byresponse monitoring component404 each of which is based on one or more events monitored byevent monitoring component402.

Categorization component

410 evaluates user responses and categories them according to applicable metrics. For instance, in a competitive boxing video game,knowledge engine220 will actively monitor the action taken by a user in response to an event (e.g., a jab from a competing avatar to the face) and continuously analyze the user's response to similar events in theinteractive environment224. The collective set of user responses to different monitored events will be correlated and categorized according to statistical techniques in order to identify the timing of responses to specific events, the types of responses to specific events and the frequency with which similar responses occur in the presence the same or similar events in theinteractive environment224. These categorized responses provide an independently searchable dataset for theknowledge engine220 in addition to the searchable associations among events and responses. Furthermore, the nature of a user response to an event will be evaluated bycategorization component410 to deduce an emotion metric associated with each response. Thus,component410 will also enable theknowledge engine220 to issue requests toaction engine222 that will enable the actions of an avatar operating in an autonomous mode to mimic the emotional character of a registered user's response to certain monitored events. The timing performance of user responses are categorized bycategorization component410 as “performance metrics,” while the type of response to specific events is categorized bycomponent410 as a “decision metric.”

Knowledge base controller

412 controls the reading and writing of “associations” represented as diverse forms of abstract data types and related operators. These associations are stored and retrieved based on the associations created bypattern analysis components408.Action engine controller414 controls the transmission of requests toaction engine222 to enable explicit actions of an avatar to be implemented by theaction engine222 based on specific monitored user responses, as monitored byresponse monitoring component404, or actions requested bypattern analysis component408 andcategorization component410 when one or more avatars in theinteractive environment224 are operating in an autonomous mode, or when a user is not actively logged-in, or taking action in response to events occurring in theinteractive environment224.

FIGS. 5A and 5B depict alternative embodiments of the software-implemented system in a client/server configuration. In particular, these figures representclient devices104 andserver devices102.FIG. 5A represents aserver device102 comprising one ormore input devices502, acommunication interface504, a read onlymemory506, astorage device518, aprocessor512, aprogram memory514 and one ormore output devices516. Each of these components of theserver device102 is communicatively coupled tocommunication bus510.Program memory514 includes aknowledge base518 and aserver knowledge engine520.Server knowledge engine520 continuously monitors and analyses user responses and forms associations between events monitored byclient knowledge engine540 in theinteractive environment536 and the user responses.Knowledge engine520 also applies processes to categorize user responses according to various metrics, including a decision metric, a performance metric and an emotion metric. Knowledge based518 stores associations between monitored events and monitored user responses and permits the organized storage of the associated events and responses according to the categories in which user responses have been categorized based on the applicable metric (i.e., decision, performance or emotion).

FIG. 5B illustrates aclient device104 comprised of one ormore input devices522, a communication interface for communicating to and fromserver device102, as illustrated in theFIG. 5A, a read onlymemory526, astorage device528, aprocessor532, aprogram memory534 and one ormore output devices530. Each of these components is coupled tocommunication bus525 to facilitate inter-component communication.Program memory534 includes aclient knowledge engine540, anaction engine542, aninteractive environment536 and anoperating system538. Each of the components included inprogram memory534 are used byprocessor532 for execution of theinteractive environment536. Theinteractive environment536 is a computer-generated environment that controls the autonomous execution of computer-generated events. In an embodiment theinteractive environment536 is a computer-generated simulation environment, while on a different embodiment theinteractive environment536 is a computer-generated video game environment.Client knowledge engine540 continuously monitors events in theinteractive environment536 and user responses to those events when a registered user is logged-in and actively responding to events in theinteractive environment536.Client knowledge engine540 compiles a profile of the user responses and actively reports those responses toserver knowledge engine520 for analysis, association and categorization. User responses are categorized by several different metrics, including a decision metric, a performance metric and an emotion metric. After categorization, the responses are stored inknowledge base518 according to one or more associations and applicable metrics.

FIG. 6 illustrates an alternative client/server embodiment for the software-implemented system. In particular, this embodiment includes one ormore server devices102a-102d (FIG. 6A) and one ormore client devices104a-104d (FIG. 6B).Server device102 includes one ormore input devices602, acommunication interface604 for sending and receiving communications with one ormore client devices104, a read onlymemory606, astorage device608, aprocessor612, aprogram memory626 and one ormore output devices610. Included inprogram memory626 are aknowledge base616, aserver knowledge engine618, and anaction engine620.Server knowledge engine618 directs the actions ofaction engine620 by issuing requests that are translated into a specific actions performed by avatars under the control of theaction engine620.Server knowledge engine618 interacts withinteractive environment622, which is executed by operatingsystem624, to monitor events occurring in thisenvironment622.Processor612 interacts withprogram memory626 for execution of operations performed byserver knowledge engine618 andaction engine620 and for rapid search, storage and retrieval of associations and rapid searching and evaluation of user responses according to categorized metrics. Each of the components in theserver device102 is communicatively coupled viacommunication bus614.

In the embodiment depicted inFIGS. 6A and 6B,server knowledge engine618 andclient knowledge engine644 actively interoperate to monitor user responses to events occurring in theinteractive environment622. User responses are monitored onclient device104 byclient knowledge engine644, which is store in alocal program memory642. As shown inFIG. 6B,client device104 includes one ormore input devices628, acommunication interface630, a read onlymemory632, astorage device634, aprocessor636,program memory642 and one ormore output device646. These components are communicatively coupled to each other overcommunication bus640. In this embodiment, users respond to events in theinteractive environment622 by taking action (e.g., moving joysticks, touching portions of a touch screen, etc.) on aninput device628 of theclient device104 and these responses are actively and continuously monitored byclient knowledge engine644. Althoughclient knowledge engine644 monitors user responses on theclient devices104, in this embodiment the record of monitored responses is communicated toserver knowledge engine618 for analysis of responses and events, the forming of associations between responses and events, and the categorization of user responses according to specific metrics. In an embodiment,interactive environment622 executes on aserver device102. In an alternative embodiment, theinteractive environment622 is executed on a plurality of servers, each having an independent though cohesive role in managing the execution of events in theinteractive environment622 to which registered users and autonomously executing avatars may respond.Server knowledge engine618 analyzes the user responses received fromclient devices104, associates each user response to a monitored event in theinteractive environment622, and categorizes user responses according to at least one decision metric, a performance metric and an emotion metric.

The metrics applied in this system represent various timing measures related to user responses (i.e., performance metrics) and various outcomes of certain user responses to monitored events in interactive environment622 (i.e., decision metrics). In addition to performance metrics and decision metrics into which various user responses may be categorized, the user responses can also be categorized according to an emotion metric. An emotion metric is based on one or more performance metrics and decision metrics and is a composite metric that is determined byserver knowledge engine618 and associated with the responses to related events. These associations are used in an autonomous mode to control the actions of one or more avatars associated with the registered user to enable the actions of the avatar to be performed in a manner consistent with past user responses of the user. In essence,server knowledge engine618 enables the active, ongoing and continuous profiling of user responses, and the association and categorization of those responses for the purpose of determining and applying actions using an avatar in an autonomous operating mode that are consistent with the responses and related emotions associated with those responses that were displayed by a user when that user was logged-in and responding to events in theinteractive environment622.

FIG. 7 illustrates a flowchart outlining a process for monitoring and controlling avatars in an interactive environment. This process commences atstep702 and proceeds with the monitoring of events in an interactive environment, as shown instep704. While monitoring events in the interactive environment, user responses to these events will be monitored and measured, as shown asstep706, and these user responses will be compiled, analyzed and categorized according to one or more performance metrics, decision metrics and emotion metrics, as shown asstep708. The categorized user responses will be stored, as shown asstep710, for later searching and use by a knowledge engine in formulating requests to control the actions of one or more avatars in the interactive environment while operating an autonomous mode. The avatars perform actions in the autonomous mode when the subscriber to whom the avatars have registered is not logged-in or actively responding to monitored events in the interactive environment. The process concludes, as shown atstep712, after the user responses are categorized and stored in a knowledge base.

FIG. 8 illustrates a process for applying a response based on previous user responses after a comparative evaluation of prior user responses. The process commences atstep802 with the monitoring of events in an interactive environment, as shown asstep804. A knowledge base storing prior user responses is searched and those responses will be evaluated and compared with a record indicated user responses to prior monitored events, as shown instep806. In searching and evaluation prior user responses in a knowledge base, the knowledge engine will determine a set of most likely alternative user responses to specific monitored events, as shown asstep808 that can be applied by one or more avatars for a registered user. The set of most likely alternative responses to each monitored event will be determined by analyzing patterns in the responses to past events in the interactive environment and comparing in real-time those past events with the monitored events. One or more pattern analysis techniques can be applied, including techniques that analyze the frequency of occurrence of specific types of responses for type of previously monitored event. Statistical relevance ranking techniques can also be applied alone or in combination with a frequency of occurrence measure to determine the most likely response, or set of responses, to monitored events. Once a most likely response is determined, the selected response will be applied in the interactive environment by one or more avatars for a registered user, as shown atstep810, and the process will come to an end, as shown atstep812. Although the term “responses” as used here applies generally to the controlled actions of registered users in an interactive environment, this term is also intended to refer to the specific actions taken by avatars in this environment in the absence of direct, controlled activity by a user.

FIG. 9 illustrates a process for monitoring user responses and determining when to enter a protected wait state. This process commences atstep902 with the monitoring of user responses (shown at step904) and the comparison of each monitored response to stored responses in a knowledge base, as shown atstep906. One or more pattern analysis methods will be applied to the monitored user responses (shown at step908) based on the comparisons performed instep906. New user responses to similar previously monitored events will be associated with these events and stored to enable the knowledge base to be updated with an active, real-time record of associations between events and responses, as shown asstep910. After updating, the interactive environment will be monitored to determine if a user has logged off, as shown instep912. If a user has not logged off, a wait time will be checked to determine whether the length of time between the occurrence of the last action performed by the user and the present time exceeds a wait time threshold for receiving a response from a user to a monitored event in the interactive environment, as shown instep914. If the wait time has been exceeded, then the one or more avatars for a registered user will enter into a protected wait state in the interactive environment and await commands or requests to take actions from the user, as shown atstep916.

Alternatively, if the user has logged off, as shown instep912, then the user's avatars will enter into the protected wait state in the interactive environment, as shown instep916. If the time in which a user takes an action has not exceeded the wait time, as shown atstep914, then the user responses will continue to be actively monitored and compared to the record of stored responses in the knowledge base, as shown atstep906. This comparative real-time analysis of newly monitored responses with previously stored monitored responses continues while the interactive environment is active in order to continually identify, associate and categorize new and different responses to events in this environment, which analysis is shown atstep908. The updating of a knowledge base with associations of new user responses to events in the interactive environment is shown instep910. If the user logs off, the protected wait state will be entered (step916) and the process will end, shown atstep918.

FIG. 10 depicts a process that enables an avatar to interact with a user when that user is not logged-in to an interactive environment. The process commences asstep1002 with the monitoring of events in an interactive environment while an avatar is in a protected wait state, as shown atstep1004. Depending on the monitored event, certain actions may be required by the avatar, as indicated bystep1006. If no action is required, the events in the interactive environment will continue to be monitored and the avatar will remain in the protected operational wait state. However, if action is required, an action alert will be sent to the, user as shown atstep1008, and the avatar await a response from the user, as shown atstep1010. If no response is received, then the avatar will enter into an autonomous operational mode, as shown atstep1012, which will enable each of the avatars of a registered user to interact with events in an interactive environment based on the collective record of associations between events and user responses which are stored in a knowledge base. The actions most likely to be applied will be determined from the knowledge base of stored associations and an action engine will be activated to explicitly control the execution of one or more actions of a user's avatar(s) based on the control rules stored in the avatar control rules database shown inFIG. 3. In this operational mode, a knowledge engine will send requests to the action engine based on the stored associations and continuously analyze the actions of avatars in response to monitored events in the interactive environment. The knowledge engine and the action engine will have exclusive control over the actions of each user's avatars in this mode. On the other hand, if a user response is received, as shown atstep1010, then the process will transition to the steps shown inFIG. 11, as discussed below.

Referring back tostep1008, an action alert can be issued by one or more of a registered users avatars in the form of electronic mail messages or telephone calls from each of the user's avatars in which a speech synthesized voice of an avatar describes the events in the interactive environment to the user. The electronic mail message, also referred to as an “Emergency Action Message,” can include a request for specific user input to enable the avatar to take action in the interactive environment in response to an imminent event or threat.

If a user responds to an action alert received on a telephone call with a command, the avatar will attempt to execute the command, as shown atstep1102 inFIG. 11. In processing the user response, a knowledge engine will determine whether an independent user response is required, as shown instep1114. An independent user response will be required if the avatar that placed the action alert cannot perform an action specified in a received user command without independent user action in the interactive environment. In such instances, the operational mode of the avatars in the interactive environment will transition to a competitive duality mode, as shown atstep1106. In this operational mode, the avatars would engage in independent actions in the interactive environment that would be directly competitive with the actions of the user once the user returned to a logged-in status in this environment. In an alternative embodiment, a group of avatars may be controlled by the user and a collaborative team of avatars can be formed that engages in competitions as a team against the group controlled by the user. The group of user controlled avatars and the collaborative teams of avatars can also respond to events in the interactive environment competitively after the user returns to a logged-in status in the interactive environment.

In the alternative, if an independent user response is not required, then the user and the user's one or more avatars would respond to events in an interactive environment in an assisted duality mode (1108) in the interactive environment. The competitive duality mode is an operating mode in which user responses in an interactive environment actively compete against one or more actions of a user's avatars in the environment. In this case, both the avatars and the user will respond to monitored events in the interactive environment and to each other's actions in that environment. In the assisted duality mode, the user and the user's avatars operate in a collaborative fashion and the user's actions will supplement the independent actions of the user's avatars while those avatars respond to monitored events in an interactive environment.

Returning toFIG. 11, if a user response (i.e., a user action in the interactive environment) is received instead of a user command (i.e., an email or verbal directive to the avatar issuing the action alert), then the interactive environment will enter a real user operational mode, as shown atstep1114, and then return, as shown atstep1116. A user response involves a user logging into the interactive environment to take control of the one or more avatars in the interactive environment and to respond directly to monitored events in that environment.

FIG. 12 illustrates a process for autonomous operation of avatars in an interactive environment. The process commences atstep1202 with the monitoring of events in an interactive environment, as shown atstep1204. In the autonomous operating mode, one or more avatars of each registered user continues to actively compete against the avatars of other registered users regardless of the user's log-in or control status of those avatars. Avatars can engage in competitions with individual avatars, or self-organize into collaborative teams of avatars for competitions against other groups of avatars or for competitions in response to monitored events in the interactive environment. In this operating mode, multiple servers can host the interactive environment to enable the avatars of all registered users to actively compete against each other for performance rankings in selected competitive events in the interactive environment. The competitive events may involve individual competition among avatars in response to actions of other avatars, competitive assessments of avatars in response to monitored events in the interactive environment, and the formation of collaborative teams of avatars which engage in competition or respond to monitored events as collaborative teams. These events may be held at prescheduled times and announced to all registered users as “World Olympic Avatar Competitions.” Thus, requests for participation in such competitions can be received by registered users, as shown atstep1206. The requests are issued by administrators of the interactive environment in which the avatar competitions will be held.

Competition among the avatars of registered users who accepted the requests will commence, as shown atstep1208, and these avatars will take actions in response to events in an autonomous operational mode. The competitions involve multiple interactive scenarios requiring competitive interaction by one or more of each registered users' avatars against other competing avatars in an interactive environment. The performance of the avatars in the interactive scenarios will be evaluated and used to determine a competitive ranking, as shown atstep1210. The actions of the avatars in these autonomous mode interactive competitions are based in significant part on the associations between events and user responses stored in a knowledge base, and the categorization of these user responses according to selected metrics. A knowledge engine categorizes the user responses while the user actively controls and executes responses to events in the interactive environment. The categorization of user responses is based on specific performance metrics, decision metrics and emotion metrics. These metrics affect how avatars will apply actions in the interactive environment based while operating in the autonomous competition mode. More specifically, the actions applied by avatars in the autonomous competition mode are based on pattern analyses that are applied to user responses by the knowledge engine when those responses are initially made to monitored events in the interactive environment, as shown atstep1212. The knowledge engine operating in conjunction with an action engine will have exclusive control over the actions of avatars in this competition mode. Registered users can have active avatars perform actions in the autonomous mode or the autonomous competition mode for an indefinite time period. In an embodiment, the indefinite period can extend beyond the lifespan of a registered user. A performance ranking of the actions of the avatars will be compiled, as shown atstep1214, and a determination is made to determine which avatar scored the highest in an autonomous and fully hosted competition in the interactive environment. After compiling performance rankings for the avatars in the autonomous competition mode, the process ends, as shown instep1216.

FIG. 13 illustrates an embodiment of a process for a competitive operational mode. In this mode, a registered user actively competes with one or more of the user's avatars in the interactive environment. This process commences atstep1302 with the monitoring of events in the interactive environment, as indicated atstep1304. User responses are monitored and measured according to various categorizes of metrics, including performance metrics, decision metrics and emotion metrics, and the responses are associated and categorized. The user responses are associated with the monitored event that caused the specific response and this association is stored in order to maintain an updated knowledge base, as shown instep1308. After updating, in the autonomous mode the knowledge base will be searched according to collective associations of responses and events, in a manner similar to the storage of information in an associated form in the human brain.

In searching these associations for user responses to monitored events, as shown atstep1310, one or more pattern analysis techniques will be applied to determine the most likely user response in the competitive operational mode. As indicated above, in the competitive duality mode a user is active in an interactive environment but is pitted against one or more of its registered avatars who take action in response to the actions of the human user as well as automated events in the interactive environment, as shown atstep1312. A knowledge engine will actively search a knowledge base of stored user responses and issue requests based on the associations formed between monitored events and prior user responses for the purpose of controlling the actions of avatars in the interactive environment while continuously monitoring events in the environment.

The actions of the avatars are based on requests made by a knowledge engine to an action engine and these requests are based on real-time searching and identification of associations of user responses and prior monitored events. These associations enable the knowledge engine to readily determine what actions are necessary in response to monitored events. The applicable action is determined from statistical relevance measures, a frequency of occurrence measure or other applicable statistical measures that are applied to determine in real-time the most likely response a user would take in response to a monitored event. These responses would be applied by controlled actions regulated by the action engine for the purpose of directing one or more avatars in the interactive environment in a competitive operational mode, as shown instep1314. After controlling avatar action based on the most likely user response, this process ends as shown atstep1316.

FIG. 14 illustrates a process for controlling the actions of an avatar in an assisted duality operational mode. At the commencement of this process, shown atstep1402, events in an interactive environment are continuously monitored, shown atstep1404, and the monitoring and measuring of user responses to these events is performed, as shown atstep1406. A knowledge base is continually updated with associations and categorizations of user responses as events in the interactive environment are continuously monitored, as shown asstep1408. The knowledge base is searched by a knowledge engine for user responses according to stored associations and categorizations, as shown atstep1410, and one or more pattern analysis methods are applied to determine the most likely user response to the monitored events in the interactive environment, which response will be applied by a user's avatars in the interactive environment in the assisted duality operational mode, shown atstep1412.

In the assisted duality operational mode, a user's avatar will monitor the actions of the user while that user actively controls the avatars in the interactive environment. In monitoring the user's actions to events in the interactive environment, a knowledge engine will continuously measure those actions, build associations between the user's actions to monitored events in the interactive environment, and categorize the actions according to performance metrics, decision metrics, and at least one emotion metric. In this context, the “actions” of a user are the “responses” made by the user to events occurring in the interactive environment. After the knowledge engine determines the most likely user response to be applied by an avatar in the assisted duality mode, an action engine will cause the avatar to execute an action, which action will based on the most likely user response to the monitored event in the interactive environment, as shown instep1414. After application of an action based on the most likely response, the process ends, as shown atstep1416.

Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that a wide variety of alternate and/or equivalent implementations may be substituted for the specific embodiments shown and described without departing from the scope of the present invention. This application is intended to cover any adaptations or variations of the embodiments discussed herein.