No description available.

The development of the research "Goal Formulation in Intelligent Agents" is anchored on the rationale that to be truly called "intelligent", an agent must not only be capable of knowing how to achieve its given goals; preferably, it must also have the capability to formulate its own goals. It must be able to detect its own goals, assess their feasibility, prioritize them, evaluate their validity as to whether they have to be acted upon, terminated, or suspended. This research has developed and implemented an intelligent system that is capable for formulating its own goals. Goal formulation refers to the intelligent behavior that an agent exhibits when reasoning about what goals to pursue and when to pursue them. It is an integrated reasoning mechanism that identifies the relevant goals that an agent needs to accomplish to affect the external world (Goal detection); constantly updates the qualitative and quantitative information attributed to the active goals as events unfold (Active goal status evaluation); assesses whether a goal is attainable through the application of the agent's own actions (Goal achievability assessment); and dynamically evaluates the relative merits of an agent's tasks, provides the agent with a sound basis to make a rational choice among a set of competing alternatives and then decides what to do next based on the choice made (Next action selection). In the development of the goal formulator, the types and structure of the required knowledge are identified; architectures for the various goal formulation components have been designed; and algorithms for the various goal formulation reasoning mechanisms (e.g. application of NPV economic decision criterion) have been developed and implemented in Prolog. To prove the applicability of the goal formulation concepts that this research had developed, the system was applied in the housekeeping domain. Simulations of some housekeeping cases are provided.

University of Technology, Sydney. Faculty of Information Technology. / This thesis establishes Virtual Institutions as a comprehensive software engineering technology for the development of 3D Virtual Worlds that require normative regulation of participants’ interactions (such as the commercially-oriented Virtual Worlds). 3D Virtual Worlds technology currently offers somewhat unregulated environments without means to enforce norms of behavior and interaction rules on their inhabitants. Furthermore, existing methodologies for Virtual Worlds development focus primarily on the design side of the “look-and-feel” of the inhabited space. Consequently, in current 3D Virtual Worlds it is difficult to keep track of the deviant behavior of participants and to guarantee a high level of security and predictable overall behavior of the system. The Virtual Institutions Methodology proposed by this dissertation is focused on designing highly secure heterogeneous Virtual Worlds (with humans and autonomous agents participating in them), where the participants behave autonomously and make their decisions freely within the limits imposed by the set of norms of the institution. It is supported by a multilayer model and representational formalisms, and the corresponding tools that facilitate rapid development of norm-governed Virtual Worlds and offer full control over stability and security issues. An important part of the Virtual Institutions Methodology is concerned with the relationship between humans and autonomous agents. In particular, the ways to achieve human-like behavior by learning such behavior from the humans themselves are investigated. It is explained how formal description of the interaction rules together with full observation of the users’ actions help to improve the human-like believability of autonomous agents in Virtual Institutions. The thesis proposes the concept of implicit training, which enables the process of teaching autonomous agents human characteristics without any explicit training efforts required from the humans, and develops the computational support for this new learning method. The benefits of using Virtual Institutions are illustrated through applying this technology to the domain of E-Commerce. It is demonstrated that providing shoppers with a normative environment that offers immersive experience and supports important real world attributes like social interaction, location awareness, advanced visualization, collaborative shopping and impulsive purchases can improve existing practices in E-Commerce portals.

Agents, by definition, (1) are situated in an environment upon which their actions affect changes and (2) have some level of autonomy from the control of humans or other agents. Being situated requires that the agent have a mechanism for sensing the environment as well as actuators for changing the environment. Autonomy implies that each agent has the freedom to make their own decisions. Rational agents are those agents that decide to execute actions that are in their “best interests” according to their desires, using a model of those desires on which they make those decisions. Action selection is complicated due to uncertainty when operating in a dynamic environment or where other actors (agents) can also influence the environment. This dissertation presents an action selection framework and algorithms that are (1) rational with respect to multiple desires and (2) responsive with respect to changing desires. Agents can use the concept of commitments, and the subsequent communication of those commitments, to coordinate their actions and reduce their uncertainty. Coordination is layered on top of this framework by describing and analyzing how commitments affect the agents’ desires in their action selection models. This research uses the domain of UAV surveillance to experimentally explore the balance between under-commitment and over-commitment. Where previous approaches concentrate on the semantics of commitment, this research concentrates on the pragmatics of commitment, describing how to use utility calculations to enable an agent to decide when making a commitment is in its best interests. / text

University of Technology, Sydney. Faculty of Information Technology. / This thesis establishes Virtual Institutions as a comprehensive software engineering technology for the development of 3D Virtual Worlds that require normative regulation of participants’ interactions (such as the commercially-oriented Virtual Worlds). 3D Virtual Worlds technology currently offers somewhat unregulated environments without means to enforce norms of behavior and interaction rules on their inhabitants. Furthermore, existing methodologies for Virtual Worlds development focus primarily on the design side of the “look-and-feel” of the inhabited space. Consequently, in current 3D Virtual Worlds it is difficult to keep track of the deviant behavior of participants and to guarantee a high level of security and predictable overall behavior of the system. The Virtual Institutions Methodology proposed by this dissertation is focused on designing highly secure heterogeneous Virtual Worlds (with humans and autonomous agents participating in them), where the participants behave autonomously and make their decisions freely within the limits imposed by the set of norms of the institution. It is supported by a multilayer model and representational formalisms, and the corresponding tools that facilitate rapid development of norm-governed Virtual Worlds and offer full control over stability and security issues. An important part of the Virtual Institutions Methodology is concerned with the relationship between humans and autonomous agents. In particular, the ways to achieve human-like behavior by learning such behavior from the humans themselves are investigated. It is explained how formal description of the interaction rules together with full observation of the users’ actions help to improve the human-like believability of autonomous agents in Virtual Institutions. The thesis proposes the concept of implicit training, which enables the process of teaching autonomous agents human characteristics without any explicit training efforts required from the humans, and develops the computational support for this new learning method. The benefits of using Virtual Institutions are illustrated through applying this technology to the domain of E-Commerce. It is demonstrated that providing shoppers with a normative environment that offers immersive experience and supports important real world attributes like social interaction, location awareness, advanced visualization, collaborative shopping and impulsive purchases can improve existing practices in E-Commerce portals.

Made available in DSpace on 2014-06-12T15:52:31Z (GMT). No. of bitstreams: 1 license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2009 / Atores Sintéticos são agentes inteligentes que possuem personalidade, atuam em ambientes multimídias ou virtuais e são representados graficamente por avatares. Estas entidades são utilizadas em sistemas de simuladores como forma de enriquecer a interação entre a aplicação e seus usuários, proporcionando assim aumento da credibilidade do projeto. Com a sua popularização sugiram vários projetos que os utilizam. Estes projetos têm diversos aspectos em comum. Por exemplo, neles seus atores se comunicam e possuem modelos de personalidade. Contudo, apesar das similaridades os projetos encontrados na literatura foram implementados de forma independente. Ou seja, tiveram seu código fonte desenvolvido do zero. Porém, no âmbito dos sistemas multiagentes tal problema não ocorre devido à existência de diversos frameworks multiagentes que agrupam as funcionalidades comuns a tais sistemas, fazendo com que o desenvolvedor implemente apenas o que é relativo ao seu contexto de simulação. Porém, estes frameworks não são utilizados em sistemas multiarores, pois não suportam as peculiaridades dos sistemas multiatores. Buscando solucionar esse problema, este trabalho apresenta o FAct (Framework for Actors), um projeto criado especificamente para auxiliar na construção sistemas multiatores. Ele focado nas características peculiares destes sistemas e tem como principal objetivo reduzir o custo e tempo de desenvolvimento para os projetos baseados nesta tecnologia. Como forma de exemplificar o uso do FAct e avaliar o seu impacto no desenvolvimento de simuladores multiatores, serão apresentados dois simuladores implementados com base no framework

Because games set in persistent virtual game worlds (VGWs) have massive numbers of players, these games need methods of characterisation for playable characters (PCs) that differ from the methods used in traditional narrative media. VGWs have a number of particularly interesting qualities. Firstly, VGWs are places where players interact with and create elements carrying narrative potential. Secondly, players add goals, motives and driving forces to the narrative potential of a VGW, which sometimes originates from the ordinary world. Thirdly, the protagonists of the world are real people, and when acting in the world their characterisation is not carried out by an author, but expressed by players characterising their PCs. How they can express themselves in ways that characterise them depend on what they can do, and how they can do it, and this characterising action potential (CAP) is defined by the game design of particular VGWs. In this thesis, two main questions are explored. Firstly, how can CAP be designed to support players in expressing consistent characters in VGWs? Secondly, how can VGWs support role-play in their rule-systems? By using iterative design, I explore the design space of CAP by building a semiautonomous agent structure, the Mind Module (MM) and apply it in five experimental prototypes where the design of CAP and other game features is derived from the MM. The term semiautonomy is used because the agent structure is designed to be used by a PC, and is thus partly controlled by the system and partly by the player. The MM models a PC’s personality as a collection of traits, maintains dynamic emotional state as a function of interactions with objects in the environment, and summarises a PC’s current emotional state in terms of ‘mood’. The MM consists of a spreading-activation network of affect nodes that are interconnected by weighted relationships. There are four types of affect node: personality trait nodes, emotion nodes, mood nodes, and sentiment nodes. The values of the nodes defining the personality traits of characters govern an individual PC’s state of mind through these weighted relationships, resulting in values characterising for a PC’s personality. The sentiment nodes constitute emotionally valenced connections between entities. For example, a PC can ‘feel’ anger toward another PC. This thesis also describes a guided paper-prototype play-test of the VGW prototype World of Minds, in which the game mechanics build upon the MM’s model of personality and emotion. In a case study of AI-based game design, lessons learned from the test are presented. The participants in the test were able to form and communicate mental models of the MM and game mechanics, validating the design and giving valuable feedback for further development. Despite the constrained scenarios presented to test players, they discovered interesting, alternative strategies, indicating that for game design the ‘mental physics’ of the MM may open up new possibilities. The results of the play-test influenced the further development of the MM as it was used in the digital VGW prototype the Pataphysic Institute. In the Pataphysic Institute the CAP of PCs is largely governed by their mood. Depending on which mood PCs are in they can cast different ‘spells’, which affect values such as mental energy, resistance and emotion in their targets. The mood also governs which ‘affective actions’ they can perform toward other PCs and what affective actions they are receptive to. By performing affective actions on each other PCs can affect each others’ emotions, which - if they are strong - may result in sentiments toward each other. PCs’ personalities govern the individual fluctuations of mood and emotions, and define which types of spell PCs can cast. Formalised social relationships such as friendships affect CAP, giving players more energy, resistance, and other benefits. PCs’ states of mind are reflected in the VGW in the form of physical manifestations that emerge if an emotion is very strong. These manifestations are entities which cast different spells on PCs in close proximity, depending on the emotions that the manifestations represent. PCs can also partake in authoring manifestations that become part of the world and the game-play in it. In the Pataphysic Institute potential story structures are governed by the relations the sentiment nodes constitute between entities.

Two on-line task re-allocation methodologies capable of re-allocating agents to tasks on-line for minimum task completion time in dynamic environments are presented herein. The first methodology, the Dynamic Nearest Neighbour (DNN) Policy, is proposed for the operation of a fleet of vehicles in a city-like application of the dial-a-ride problem. The second methodology, the Dynamic Re-Pairing Methodology (DRPM) is proposed for the interception of a group of mobile targets by a dynamic team of robotic pursuers, where the targets are assumed to be highly maneuverable with a priori unknown, but real-time trackable, motion trajectories. Extensive simulations and experiments have verified the DNN policy to be tangibly superior to the first-come-first-served and nearest neighbour policies in minimizing customer mean system time, and the DRPM to be tangibly efficient in the optimal dynamic re-pairing of multiple mobile pursuers to multiple mobile targets for minimum total interception time.

Two on-line task re-allocation methodologies capable of re-allocating agents to tasks on-line for minimum task completion time in dynamic environments are presented herein. The first methodology, the Dynamic Nearest Neighbour (DNN) Policy, is proposed for the operation of a fleet of vehicles in a city-like application of the dial-a-ride problem. The second methodology, the Dynamic Re-Pairing Methodology (DRPM) is proposed for the interception of a group of mobile targets by a dynamic team of robotic pursuers, where the targets are assumed to be highly maneuverable with a priori unknown, but real-time trackable, motion trajectories. Extensive simulations and experiments have verified the DNN policy to be tangibly superior to the first-come-first-served and nearest neighbour policies in minimizing customer mean system time, and the DRPM to be tangibly efficient in the optimal dynamic re-pairing of multiple mobile pursuers to multiple mobile targets for minimum total interception time.

<p>This thesis investigates the possibilities of explaining higher cognition as internal simulations of perception and action at an abstract level. Relatively recent findings in both neuroscience and psychology indicates that both perception and action can be internally simulated by activating sensory and motor areas in the brain in absence of sensory input and without any resulting overt behavior. An investigation was conducted in order to test the hypothesis that perception can be simulated in a mobile robot using abstractions. The result from this investigation showed that this was indeed the case but that the accuracy was limited. The simulations allowed the robot to anticipate long chains of future situations but were not good enough to support any overt behavior. To further improve the results there is a need for better training techniques and/or a more complex architecture.</p>