Published Article / To enable meaningful and consistent communication between different software systems in a particular domain (such as manufacturing, law or medicine), a standardised vocabulary and communication language is required by all the systems involved. Concepts in the domain about which the systems want to communicate are formalized in an ontology by establishing the meaning of concepts and creating relationships between them. The inputs to this process in found by analysing the physical domain and its processes. The resulting ontology structure is a computer useable representation of the physical domain about which the systems want to communicate. To enable the long term persistence of the actual data contained in these concepts and the enforcement of various business rules, a sufficiently powerful database system is required. This paper presents the design of a manufacturing ontology and its accompanying relational database schema that will be used in a manufacturing test domain.

This thesis proposes a holistic and hierarchical architecture to Multi-agent System design, in order to resolve the collaboration problem in diabetic healthcare system. A diabetic healthcare system is a complex and social system in the case that it involves many actors and interrelations. Collaborations among various healthcare actors are vital to the quality of diabetic healthcare. The collaboration problem is manifested by the problems of accessibility and interoperability. To support the collaboration in diabetic healthcare as such a complex and social system, the MAS must have corresponding social entities and relationships. Therefore, it is assumed that theories explaining social activity can be applied to design of MAS. Activity Theory, specifically its holistic triangle model from Engström and hierarchy thinking, provides theoretical supports to the design of individual agent architecture and MAS coordination mechanism. It is argued that the holistic and hierarchical aspects should be designed in a MAS when applied to the healthcare setting. The diabetic healthcare system is analyzed on three levels based on the hierarchy thinking. The collaboration problem is analyzed and resolved via MAS coordination. Based on the holistic activity model in Activity Theory, Müller’s Vertical Layered Architecture is re-conceptualized in the Control Unit and Knowledge Base design. It is also argued that autonomy, adaptivity and persona should be especially focused when designing the interaction between an agent system and human users. This study has firstly identified some important social aspects and the technical feasibility of embedding those identified social aspects in agent architecture design. Secondly, a MAS was developed to illustrate how to apply the proposed architecture to design a MAS to resolve the collaboration problem in diabetic healthcare system. We have designed and implemented an agent system – IMAS (Integrated Multi-agent System) to validate the research questions and contributions. IMAS system provides real time monitoring, diabetic healthcare management and decision supports to the diabetic healthcare actors. A user assessment has been conducted to validate that the quality of the current diabetic healthcare system can be improved with the introduction of IMAS.

Abstract The use of agent systems has increased enormously, especially in the field of mobile services. Intelligent services have also increased rapidly in the web. In this thesis, the utilization of software agent technology in mobile services and decentralized intelligent services in the multimedia business is introduced and described. Both Genie Agent Architecture (GAA) and Decentralized International and Intelligent Software Architecture (DIISA) are described. The common problems in decentralized software systems are lack of intelligence, communication of software modules and system learning. Another problem is the personalization of users and services. A third problem is the matching of users and service characteristics in web application level in a non-linear way. In this case it means that web services follow human steps and are capable of learning from human inputs and their characteristics in an intelligent way. This third problem is addressed in this thesis and solutions are presented with two intelligent software architectures and services. The solutions of the thesis are based on a combination of neural network and agent technology. To be more specific, solutions are based on an intelligent agent which uses certain black box information like Self-Organized Map (SOM). This process is as follows; information agents collect information from different sources like the web, databases, users, other software agents and the environment. Information is filtered and adapted for input vectors. Maps are created from a data entry of an SOM. Using maps is very simple, input forms are completed by users (automatically or manually) or user agents. Input vectors are formed again and sent to a certain map. The map gives several outputs which are passed through specific algorithms. This information is passed to an intelligent agent. The needs for web intelligence and knowledge representation serving users is a current issue in many business solutions. The main goal is to enable this by means of autonomous agents which communicate with each other using an agent communication language and with users using their native languages via several communication channels.

No description available.

Agent-based systems usually operate in real-time, stochastic and dynamic environments. Many theoretical and applied techniques have been applied to the investigation of agent architecture with respect to communication, cooperation, and learning, in order to provide a framework for implementing artificial intelligence and computing techniques. Intelligent agents are required to be able to adapt and learn in uncertain environments via communication and collaboration (in both competitive and cooperative situations). The ability of reasoning and learning is one fundamental feature for intelligent agents. Due to the inherent complexity, however, it is difficult to verify the properties of the complex and dynamic environments a priori. Since analytic techniques are inadequate for solving these problems, reinforcement learning (RL) has appeared as a popular approach by mapping states to actions, so as to maximise the long-term rewards. Computer simulation is needed to replicate an experiment for testing and verifying the efficiency of simulation-based optimisation techniques. In doing so, a simulation testbed called robot soccer is used to test the learning algorithms in the specified scenarios. This research involves the investigation of simulation-based optimisation techniques in agent-based systems. Firstly, a hybrid agent teaming framework is presented for investigating agent team architecture, learning abilities, and other specific behaviors. Secondly, the novel reinforcement learning algorithms to verify goal-oriented agents; competitive and cooperative learning abilities for decision-making are developed. In addition, the function approximation technique known as tile coding (TC), is used to avoid the state space growing exponentially with the curse of dimensionality. Thirdly, the underlying mechanism of eligibility traces is analysed in terms of on-policy algorithm and off-policy algorithm, accumulating traces and replacing traces. Fourthly, the "design of experiment" techniques, such as Simulated Annealing method and Response Surface methodology, are integrated with reinforcement learning techniques to enhance the performance. Fifthly, a methodology is proposed to find the optimal parameter values to improve convergence and efficiency of the learning algorithms. Finally, this thesis provides a serious full-fledged numerical analysis on the efficiency of various RL techniques.

Made available in DSpace on 2014-12-17T15:47:59Z (GMT). No. of bitstreams: 1 MarcilioOM_DISSERT.pdf: 2842138 bytes, checksum: a048e635c669d6652fa350ab1d82eadc (MD5) Previous issue date: 2012-09-05 / There is a need for multi-agent system designers in determining the quality of systems in the earliest phases of the development process. The architectures of the agents are also part of the design of these systems, and therefore also need to have their quality evaluated. Motivated by the important role that emotions play in our daily lives, embodied agents researchers have aimed to create agents capable of producing affective and natural interaction with users that produces a beneficial or desirable result. For this, several studies proposing architectures of agents with emotions arose without the accompaniment of appropriate methods for the assessment of these architectures. The objective of this study is to propose a methodology for evaluating architectures emotional agents, which evaluates the quality attributes of the design of architectures, in addition to evaluation of human-computer interaction, the effects on the subjective experience of users of applications that implement it. The methodology is based on a model of well-defined metrics. In assessing the quality of architectural design, the attributes assessed are: extensibility, modularity and complexity. In assessing the effects on users' subjective experience, which involves the implementation of the architecture in an application and we suggest to be the domain of computer games, the metrics are: enjoyment, felt support, warm, caring, trust, cooperation, intelligence, interestingness, naturalness of emotional reactions, believabiliy, reducing of frustration and likeability, and the average time and average attempts. We experimented with this approach and evaluate five architectures emotional agents: BDIE, DETT, Camurra-Coglio, EBDI, Emotional-BDI. Two of the architectures, BDIE and EBDI, were implemented in a version of the game Minesweeper and evaluated for human-computer interaction. In the results, DETT stood out with the best architectural design. Users who have played the version of the game with emotional agents performed better than those who played without agents. In assessing the subjective experience of users, the differences between the architectures were insignificant / H? uma necessidade dos projetistas de sistemas multiagentes em determinar a qualidade dos sistemas logo nas fases iniciais do processo de desenvolvimento. As arquiteturas dos agentes tamb?m fazem parte do projeto desses sistemas, e por isso tamb?m precisam ter sua qualidade avaliada. Motivados pelo importante papel que as emo??es desempenham em nossas vidas di?rias, pesquisadores de agentes incorporados t?m buscado criar agentes capazes de ter intera??o afetiva e natural com usu?rios, de forma que produza algum resultado ben?fico ou desej?vel. Para isso, v?rios estudos propondo arquiteturas de agentes com emo??es foram surgindo sem o acompanhamento dos devidos m?todos necess?rios para avalia??o dessas arquiteturas. Assim, o objetivo principal deste trabalho ? propor uma metodologia de avalia??o de arquiteturas de agentes emocionais, que avalia os atributos de qualidade do projeto das arquiteturas, al?m da avalia??o da intera??o humano computador, sobre os efeitos na experi?ncia subjetiva dos usu?rios de aplica??es que as implementam. A metodologia ? baseada em um modelo de m?tricas bem definido. Na avalia??o da qualidade do projeto arquitetural, os atributos avaliados s?o: extensibilidade, modularidade e complexidade. Na avalia??o dos efeitos na experi?ncia subjetiva dos usu?rios, que envolve a implementa??o das arquiteturas em uma aplica??o e que sugerimos ser no dom?nio de jogos de computador, as m?tricas s?o: agradabilidade, ajuda percebida, anima??o, carinho percebido, confian?a, coopera??o, intelig?ncia, interesse, naturalidade nas rea??es emocionais, realidade do comportamento, redu??o de frustra??o e simpatia, al?m do tempo m?dio e m?dia de tentativas. Realizamos experimentos com esta metodologia e avaliamos cinco arquiteturas de agentes emocionais: BDIE, DETT, Camurri-Coglio, EBDI, Emotional- BDI. Duas das arquiteturas, BDIE e EBDI, foram implementadas numa vers?o do jogo Campo Minado e avaliadas quanto ? intera??o humano computador. Nos resultados, a DETT se destacou com o melhor projeto arquitetural. Os usu?rios que jogaram a vers?o do jogo com agentes emocionais tiveram um melhor desempenho do que os que jogaram sem agentes. Na avalia??o da experi?ncia subjetiva dos usu?rios, as diferen?as entre as arquiteturas foram pouco relevantes.

Presently, most spacecraft are controlled from ground involving activities such as up-linking the schedule of daily operations and monitoring health parameters. These activities lead to a cognitive overload on human operators. Imaging/science opportunities are lost, if any discrepancies occur during the execution of pre-planned sequences. Consequently, advanced space exploration systems for future needs demand on-board intelligence and autonomy. This thesis attempts to solve the problem of providing an adequate degree of autonomy in future generation of spacecraft. The autonomous spacecraft accept high-level goals from users and make decisions on-board to generate detailed command schedules satisfying stringent constraints posed by the harsh environment of the space, visibility criteria and scarce on-board resources. They reconfigure themselves in case of any failure and re-plan when needed. Autonomy concepts are derived in the context of complex systems by drawing analogy to living organisms and social organisations. A general autonomy framework may be defined with a six level structure comprising of the following capabilities -reflexes, awareness, self-regulation, self-healing, self-adaptation and self-evolution. A generic and reusable software architecture is proposed using hybrid multi-agent systems, which are arranged in a hierarchical manner using two types of decomposition viz. stratum and layer. The software architecture of the autonomous spacecraft is modeled as a stratified agent with a deliberative stratum, which achieves adaptive behaviour and a reactive stratum, which achieves reactive behaviour. Each individual agent has a generic structure comprising of perception, action, communication and knowledge components. It achieves the specialist capability through model-based reasoning. The knowledge models encompass: Planning knowledge describing higher-level goals, task structure and method of achieving the goals, Control knowledge encompassing the static and dynamic models of the spacecraft and Diagnostic knowledge incorporating the cause-effect relationships. The deliberative stratum is capable of planning in different time horizons and is, in turn, organised into a hierarchical agent system with three layers corresponding to different time horizons. It is composed of a long-term, medium-term and short-term planning agents, focusing on strategic issues, spacecraft level resources and specific spacecraft states respectively. The power of Petri nets is exploited for knowledge modeling as well as for plan representation. The ability of Petri nets to represent causality, concurrency and conflict relations explicitly makes it an excellent tool for representing the planning problem. Hierarchical Timed Petri Net is chosen for our modeling, since it captures the temporal requirements of the real-time spacecraft operations as well as facilitates the modeling of the system with multiple levels of abstraction. The necessary primitives for the plan representation are defined. In hierarchical modeling using Petri nets, refinement is done by a compound (high-level) transition. A compound transition models either a complex activity, which corresponds to high-level operation on spacecraft or a method, which corresponds to the agent capability. At the lowest layer, a transition in the plan represents a primitive command to the spacecraft, such as ‘switch on camera’. The Petri net unfolding technique, which is a partial order approach, is applied to derive the plans from the dynamic knowledge models. This tackles the problem of combinatorial explosion. A hierarchical planning approach is followed, in which the abstract plan is recursively decomposed using the unfolding technique and refined by way of exercising the appropriate decisions in each layer. The reactive stratum is configured with three peer level agents. The control agent executes the command schedule and has the capability for reflex action. Structural properties of Petri nets are exploited by the execution-monitoring agent and the diagnostic agent for system level diagnosis. Fault tree method is applied for fine granularity diagnosis. The resultant architecture is a cost-effective solution, since it permits reusability of knowledge models across similar missions. The knowledge models are formally verified for ensuring the absence of deadlocks, buffer overflows, recoverability and detection of unreachable modules using Petri net properties such as reachability, liveness, boundedness, safeness, reversibility and home state. The high-risk components are subjected to safety property verification, which makes the system rugged. The hierarchical composition of Petri net models (which are independently verified), preserves liveness and boundedness characteristics and thus ensuring the reliability of the integrated models. This, in turn, ensures that reliable plans are generated on-board using these good quality models. The models of the system components viz. partial order plan, conditional plan, dynamic world model, reflex model, resource model and the hierarchical models are developed and demonstrated using HPSIM and Moses Tool Suite, using examples from spacecraft domain. The long-term planning agent, with hierarchical world models, for handling high-level goals is developed and simulated using Moses Tool Suite. The plan generation using unfolding approach is demonstrated using VIPTool, which has the partial order analysis capability. In summary, the main contributions include (a) Definition of a general framework for spacecraft autonomy; (b) design of a generic and reusable architecture for autonomous spacecraft using hybrid multi-agent concepts; (c) unified knowledge representation and reasoning using Petri nets across various strata/layers; (d) application of Petri net unfolding technique in a hierarchical manner for plan generation; (e) use of structural properties of Petri nets for fault identification and location; (f) verification and validation of Petri net models using Petri net properties and (g) simulation and demonstration of the system components viz. partial order plan, conditional plan, dynamic world model, reflex model, resource model and hierarchical models, by developing examples from spacecraft domain, using HPSIM and Moses Tool Suite and demonstration of plan generation using unfolding technique using VIPTool.

In recent years many people have built Personal Assistant Agents, Information Agents and the like, and have simply added them to the operating system as auxiliary applications, without regard to architecture. This thesis argues that an agent architecture, one designed as a sophisticated representation of an individual user, should be embedded deep in the device system software, with at least equal status to the GUI – the graphical user interface. A sophisticated model of the user is then built, drawing upon contemporary Analytical Psychology – the Psychology of Subselves. The Shadowboard Agent architecture is then built upon that user model, drawing both structural and computational implications from the underlying psychology. An XML DTD file named Shadowboard.dtd is declared as a practical manifestation of the semantics of Shadowboard. An implementation of the Shadowboard system is mapped out, via a planned conversion of two existing integrated systems: SlimWinX, an event-driven GUI system; and XSpaces, an object-oriented tuplespace system with Blackboard-like features. The decision making mechanism passes logic terms and contraints between the various sub-agent components (some of which take on the role of Constraint Solvers), giving this agent system some characteristics of a Generalised Constraint Solver. A Shadowboard agent (built using the system) consists of a central controlling autonomous agent named the Aware Ego Agent, and any number of sub-agents, which collectively form an integrated but singular whole agent modelled on the user called the Digital Self. One such whole-agent is defined in a file named DigitalSelf.xml – which conforms to the schema in Shadowboard.dtd - which offers a comprehensive and generic representation of a user’s stance in a 24x7 network, in particular - the Internet. Numerous types of Shadowboard sub-agents are declared.

Abstract The Internet of Things vision proposes a global platform in an unforeseen scale for distributed applications that rely on data provided by interconnected resource-constrained things. In such large-scale systems, centralized control of system operation by a single component through vertical interactions becomes unfeasible. Ideally, decentralized control in the proximity of things enables to take into account the local dynamic resource availability and environmental characteristics that are used to optimize the application execution. To realize decentralization, capabilities for horizontal interactions, that complement the vertical interactions, and for opportunistic participation of things are needed. This thesis explores mobile agent technology to implement distributed Internet of Things applications that benefit from both vertical and horizontal interactions of the application components. First, a resource-oriented reactive mobile agent architecture and a software framework are presented. The framework facilitates RESTful interactions between agents and other system components and provides a REST-based interface to build opportunistic agent-based applications. Two agent platforms are presented that integrate resource-constrained things into the framework as mobile agent hosts. Second, mobile agent based mobile crowdsensing and edge computing applications are evaluated with large-scale simulations and real-world experiments. The results show that energy consumption is decreased in the participating things, in comparison with the existing approaches, by agent-based in-network data processing and control of the thing operation. This thesis makes a valuable contribution by enabling mobile agent operations in a heterogeneous set of resource-constrained things. The presented empirical evidence shows how mobile agent technology increases energy efficiency in distributed application execution. The presented mobile agent architecture and software framework potentially accelerate the utilization of mobile agent technology in the Internet of Things. / Tiivistelmä Esineiden Internet liittää resurssirajoitteiset sulautetut laitteet laajamittaisesti Internet-verkkoon, jossa niiden tuottamaa tietoa hyödynnetään hajautettujen järjestelmien sovelluksissa. Esineiden Internetin järjestelmien odotetaan skaalautuvan niin laajoiksi, ettei keskitetty, vertikaaliseen vuorovaikutukseen perustuva järjestelmähallinta ole enää käyttökelpoinen ratkaisu. Hajautettu hallinta lähellä tietoa tuottavia laitteita tuo etuja, kun paikallisesti sovelluksen suorituksessa otetaan huomioon toimintaympäristön tila ja voidaan reagoida dynaamisesti resurssien saatavuuteen. Hajautus Esineiden Internetin sovellusalustoilla edellyttää menetelmiä sekä laitteiden vertikaaliseen ja horisontaaliseen vuorovaikutukseen, että niiden dynaamisen osallistumisen mahdollistamiseksi sovelluksen suorittamisessa. Tässä työssä tutkittiin mobiiliagentti-teknologiaa hajautettujen sovellusten toteuttamiseen Esineiden Internet-ympäristössä. Työssä esitettiin resurssisuuntautunut arkkitehtuuri reaktiivisille mobiiliagenteille sekä ohjelmistokehys, joita käyttäen voidaan toteuttaa agenttipohjaisia hajautettuja sovelluksia. Ohjelmistokehys perustuu REST-arkkitehtuurimalliin, jonka pohjalta esitettiin yhdenmukainen rajapinta agenttien ja järjestelmäkomponenttien väliseen vuorovaikutukseen. Ohjelmistokehyksen osana toteutettiin kaksi mobiiliagentti-ohjelmistoalustaa resurssirajoitteisille sulautetuille laitteille. Työssä tarkasteltiin mobiiliagentti-pohjaisten Esineiden Internet-sovellusten energiatehokkuutta simuloinneilla sekä tosielämän kokeilla. Tarkastelun kohteeksi valittiin joukkoistettu mobiilihavainnointi sekä reunalaskennan ulottaminen resurssirajoitteisiin laitteisiin. Saadut tulokset osoittavat, että laitteiden energiankulutusta voidaan pienentää verrattuna aiemmin esitettyihin ratkaisuihin hyödyntämällä mobiiliagentteja tiedonkäsittelyyn laitteessa sekä laitteen toiminnan ohjaamiseen. Työssä esitetty resurssisuuntautunut mobiiliagenttiarkkitehtuuri sekä ohjelmistokehys edesauttavat mobiiliagentti-teknologian hyödyntämistä resurssirajoitteisissa sulautetuissa laitteissa. Tosielämän kokeista saadut tulokset osoittavat mobiiliagenttiteknologian hyötyjä hajautettujen sovellusten toteuttamisessa Esineiden Internetiin.

Simulation has been used to model combat for a long time. Recently, it has been accepted that combat is a complex adaptive system (CAS). Multi-agent systems (MAS) are also considered as a powerful modelling and development environment to simulate combat. Agent-based distillations (ABD) - proposed by the US Marine Corp - are a type of MAS used mainly by the military for exploring large scenario spaces. ABDs that facilitated the analysis and understanding of combat include: ISAAC, EINSTein, MANA, CROCADILE and BactoWars. With new concepts such as networked forces, previous ABDs can implicitly simulate a networked force. However, the architectures of these systems limit the potential advantages gained from the use of networks. In this thesis, a novel network centric multi-agent architecture (NCMAA) is pro-posed, based purely on network theory and CAS. In NCMAA, each relationship and interaction is modelled as a network, with the entities or agents as the nodes. NCMAA offers the following advantages: 1. An explicit model of interactions/relationships: it facilitates the analysis of the role of interactions/relationships in simulations; 2. A mechanism to capture the interaction or influence between networks; 3. A formal real-time reasoning framework at the network level in ABDs: it interprets the emergent behaviours online. For a long time, it has been believed that it is hard in CAS to reason about emerging phenomena. In this thesis, I show that despite being almost impossible to reason about the behaviour of the system by looking at the components alone because of high nonlinearity, it is possible to reason about emerging phenomena by looking at the network level. This is undertaken through analysing network dynamics, where I provide an English-like reasoning log to explain the simulation. Two implementations of a new land-combat system called the Warfare Intelligent System for Dynamic Optimization of Missions (WISDOM) are presented. WISDOM-I is built based on the same principles as those in existing ABDs while WISDOM-II is built based on NCMAA. The unique features of WISDOM-II include: 1. A real-time network analysis toolbox: it captures patterns while interaction is evolving during the simulation; 2. Flexible C3 (command, control and communication) models; I 3. Integration of tactics with strategies: the tactical decisions are guided by the strategic planning; 4. A model of recovery: it allows users to study the role of recovery capability and resources; 5. Real-time visualization of all possible information: it allows users to intervene during the simulation to steer it differently in human-in-the-loop simulations. A comparison between the fitness landscapes of WISDOM-I and II reveals similarities and differences, which emphasise the importance and role of the networked architecture and the addition of strategic planning. Lastly but not least, WISDOM-II is used in an experiment with two setups, with and without strategic planning in different urban terrains. When the strategic planning was removed, conclusions were similar to traditional ABDs but were very different when the system ran with strategic planning. As such, I show that results obtained from traditional ABDs - where rational group planning is not considered - can be misleading. Finally, the thesis tests and demonstrates the role of communication in urban ter-rains. As future warfighting concepts tend to focus on asymmetric warfare in urban environments, it was vital to test the role of networked forces in these environments. I demonstrate that there is a phase transition in a number of situations where highly dense urban terrains may lead to similar outcomes as open terrains, while medium to light dense urban terrains have different dynamics