Automation has become increasingly prevalent in all forms of society. Activities that are too difficult for a human or to dangerous can be done by machines which do not share those downsides. In addition, tasks can be scheduled more precisely and accurately. Increases in the autonomy have allowed for a new level of tasks which are completed by teams of automated agents rather than a single one, called cooperative control. This has many benefits; but comes at the cost of increased complexity and coordination. The main thrust of research in this field is problem based, considering communication issues as a secondary feature. There is a gap considering problems in which many changes occur as rapidly as communication and the issues that arise as a result. This is the main motivation. This research presents an approach to cooperative control in highly variable systems and tackles some of the issues present in such a system. One of the most important issues is the communication network itself, which is used as an indicator for how healthy the system is an how well it may react to future changes. Therefore using the network as an input to control allows the system to navigate between conservative and aggressive techniques to improve performance while still maintaining robustness. Results are based on a test bed designed to simulate a wide variety of problem types based on: network type; numbers of actors; frequency of changes; impact of changes and method of change. The developed control method is compared to the baseline case ignoring cooperation as well as an idealized case assuming perfect system knowledge. The baseline represents sacrifices coordination to achieve a high level of robustness at reduced performance while the idealized case represents the best possible performance. The control techniques developed give a performance at least as good as the baseline case if not better for all simulations.

Thesis (Ph.D.)--University of Wollongong, 2005. / Typescript. Includes bibliographical references: leaf 88-95.

Thesis (M.Geom.M.)--University of Melbourne, Dept. of Geomatics, 2007. / Typescript. Includes bibliographical references (leaves 100-104).

The need for modern enterprises and Web users to simply and rapidly develop and deliver platform-independent services to be accessed over the Web by the global community is growing. This is self-evident, when one considers the omnipresence of electronic services (e-services) on the Web. Accordingly, the Service-Oriented Architecture (SOA) is commonly considered as one of the de facto standards for the provisioning of heterogeneous business functionalities on the Web. As the basis for SOA, Web Services (WS) are commonly preferred, particularly because of their ability to facilitate the integration of heterogeneous systems. However, WS only focus on syntactic descriptions when describing the functional and behavioural aspects of services. This makes it a challenge for services to be automatically discovered, selected, composed, invoked, and executed – without any human intervention. Consequently, Semantic Web Services (SWS) are emerging to deal with such a challenge. SWS represent the convergence of Semantic Web (SW) and WS concepts, in order to enable Web services that can be automatically processed and understood by machines operating with limited or no user intervention. At present, research efforts within the SWS domain are mainly concentrated on semantic services automation aspects, such as discovery, matching, selection, composition, invocation, and execution. Moreover, extensive research has been conducted on the conceptual models and formal languages used in constructing semantic services. However, in terms of the engineering of semantic services, a number of challenges are still prevalent, as demonstrated by the lack of development and use of semantic services in real-world settings. The lack of development and use could be attributed to a number of challenges, such as complex semantic services enabling technologies, leading to a steep learning curve for service developers; lack of unified service platforms for guiding and supporting simple and rapid engineering of semantic services, and the limited integration of semantic technologies with mature service-oriented technologies. vi In addition, a combination of isolated software tools is normally used to engineer semantic services. This could, however, lead to undesirable consequences, such as prolonged service development times, high service development costs, lack of services re-use, and the lack of semantics interoperability, reliability, and re-usability. Furthermore, available software platforms do not support the creation of semantic services that are intelligent beyond the application of semantic descriptions, as envisaged for the next generation of services, where the connection of knowledge is of core importance. In addressing some of the challenges highlighted, this research study adopted a qualitative research approach with the main focus on conceptual modelling. The main contribution of this study is thus a framework called iSemServ to simplify and accelerate the process of engineering intelligent semantic services. The framework has been modelled and developed, based on the principles of simplicity, rapidity, and intelligence. The key contributions of the proposed framework are: (1) An end-to-end and unified approach of engineering intelligent semantic services, thereby enabling service engineers to use one platform to realize all the modules comprising such services; (2) proposal of a model-driven approach that enables the average and expert service engineers to focus on developing intelligent semantic services in a structured, extensible, and platform-independent manner. Thereby increasing developers’ productivity and minimizing development and maintenance costs; (3) complexity hiding through the exploitation of template and rule-based automatic code generators, supporting different service architectural styles and semantic models; and (4) intelligence wrapping of services at message and knowledge levels, for the purposes of automatically processing semantic service requests, responses and reasoning over domain ontologies and semantic descriptions by keeping user intervention at a minimum. The framework was designed by following a model-driven approach and implemented using the Eclipse platform. It was evaluated using practical use case scenarios, comparative analysis, and performance and scalability experiments. In conclusion, the iSemServ framework is considered appropriate for dealing with the complexities and restrictions involved in engineering intelligent semantic services, especially because the amount of time required to generate intelligent semantic vii services using the proposed framework is smaller compared with the time that the service engineer would need to manually generate all the different artefacts comprising an intelligent semantic service. Keywords: Intelligent semantic services, Web services, Ontologies, Intelligent agents, Service engineering, Model-driven techniques, iSemServ framework. / Computing / D. Phil. (Computer science)

Em cenários onde agentes inteligentes atuam e percebem dados de ambientes com cada vez mais informações, identificar somente as percepções relevantes aos objetivos pode ser crucial para que o ciclo de raciocínio do programa seja realizado em tempo hábil. Como uma solução a este problema, o presente trabalho criou um modelo de filtragem de percepções baseado no modelo DBR (Data-oriented Belief Revision) para ser aplicado em agentes BDI (Belief Desire Intention). Para isto, o trabalho estendeu e formalizou parte dos conceitos do modelo DBR tornando-o computacionalmente aplicável. Entre as contribuições deste trabalho estão a extensão da definição dos processos de Foco (seleção de dados percebidos) e Esquecimento de dados inativos; a definição e formalização de modelos de cálculo da Relevância de percepções, que permitem filtrar ou descartar dados em função dos planos do agente e de suas valorações de importância; e as definições dos modelos de armazenamento de Dados Inativos capazes de suportar diferentes cenários de utilização de agentes BDI. O resultado foi um filtro de percepções, genérico e automatizado, orientado aos objetivos de agentes BDI. Para operacionalizar o modelo criado, ele foi implementado na plataforma de desenvolvimento de agentes Jason. Análises empíricas foram realizadas para avaliar a corretude e os impactos relacionados ao tempo de pro- cessamento após a aplicação do modelo. Os experimentos realizados indicaram que o modelo de filtragem de percepções proposto neste trabalho contribui no desempenho computacional de agentes expostos a ambientes com muitos ruídos. / In scenarios where intelligent agents act and perceive data of environments with much information, identify only perceptions relevant to goals can be crucial for the agent reasoning cycle to be performed in time. As a solution to this problem, this work creates a model to filter perceptions based on DBR (Data-oriented Belief Revision) model to be applied at BDI (Belief Desire Intention) agents. In order to do it, this work has extended and formalized some of the DBR model concepts making it applicable in computer programs. Among this work contributions are the extension and definition of the processes Focus (selection of perceived data) and Oblivion of inactive data; definition and formalization of perception Relevance models, calculations that allow to filter or discard data based on agent plans and their importance values;definition of Inactive Data storage models able to support different usage scenarios of BDI agents. The result was a generic and automated perception filter oriented to the goals of BDI agents. To opera- tionalize the model, it was implemented in the agent development plataform Jason. Empirical analysis have been done to assess the correctness and identify the impact on the processing time after the model application. The results indicate that the perception filtering model proposed in this work contributes to the computational performance of agents exposed to environments with much noise.

Em cenários onde agentes inteligentes atuam e percebem dados de ambientes com cada vez mais informações, identificar somente as percepções relevantes aos objetivos pode ser crucial para que o ciclo de raciocínio do programa seja realizado em tempo hábil. Como uma solução a este problema, o presente trabalho criou um modelo de filtragem de percepções baseado no modelo DBR (Data-oriented Belief Revision) para ser aplicado em agentes BDI (Belief Desire Intention). Para isto, o trabalho estendeu e formalizou parte dos conceitos do modelo DBR tornando-o computacionalmente aplicável. Entre as contribuições deste trabalho estão a extensão da definição dos processos de Foco (seleção de dados percebidos) e Esquecimento de dados inativos; a definição e formalização de modelos de cálculo da Relevância de percepções, que permitem filtrar ou descartar dados em função dos planos do agente e de suas valorações de importância; e as definições dos modelos de armazenamento de Dados Inativos capazes de suportar diferentes cenários de utilização de agentes BDI. O resultado foi um filtro de percepções, genérico e automatizado, orientado aos objetivos de agentes BDI. Para operacionalizar o modelo criado, ele foi implementado na plataforma de desenvolvimento de agentes Jason. Análises empíricas foram realizadas para avaliar a corretude e os impactos relacionados ao tempo de pro- cessamento após a aplicação do modelo. Os experimentos realizados indicaram que o modelo de filtragem de percepções proposto neste trabalho contribui no desempenho computacional de agentes expostos a ambientes com muitos ruídos. / In scenarios where intelligent agents act and perceive data of environments with much information, identify only perceptions relevant to goals can be crucial for the agent reasoning cycle to be performed in time. As a solution to this problem, this work creates a model to filter perceptions based on DBR (Data-oriented Belief Revision) model to be applied at BDI (Belief Desire Intention) agents. In order to do it, this work has extended and formalized some of the DBR model concepts making it applicable in computer programs. Among this work contributions are the extension and definition of the processes Focus (selection of perceived data) and Oblivion of inactive data; definition and formalization of perception Relevance models, calculations that allow to filter or discard data based on agent plans and their importance values;definition of Inactive Data storage models able to support different usage scenarios of BDI agents. The result was a generic and automated perception filter oriented to the goals of BDI agents. To opera- tionalize the model, it was implemented in the agent development plataform Jason. Empirical analysis have been done to assess the correctness and identify the impact on the processing time after the model application. The results indicate that the perception filtering model proposed in this work contributes to the computational performance of agents exposed to environments with much noise.

Orientador: Ricardo Ribeiro Gudwin / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia e de Computação / Made available in DSpace on 2018-08-14T22:03:33Z (GMT). No. of bitstreams: 1 Silva_RicardoCapitanioMartinsda_M.pdf: 6323710 bytes, checksum: efe31176ea68444cc98c3b8ca83d7ad5 (MD5) Previous issue date: 2009 / Resumo: Há muito tempo, a consciência humana tem desafiado cientistas de várias áreas do conhecimento. Em computação, na última década, tem se observado um crescimento significativo dos estudos sobre consciência artificial. Uma abordagem computacional promissora é a da arquitetura Baars-Franklin desenvolvida por Stan Franklin e inspirada na teoria do workspace global de Bernard Baars. Esse trabalho visa examinar essa tecnologia e verificar as vantagens e desvantagens do seu uso na implementação de agentes inteligentes. Como o estudo de consciência artificial é inerentemente multidisciplinar, inicialmente apresentam-se as principais teorias gerais de consciência. Em seguida, é feito um levantamento dos principais trabalhos na área de consciência artificial. Por fim, utiliza-se a arquitetura Baars-Franklin no controle de uma criatura virtual em um problema de navegação autônoma. Esta dissertação traz uma série de contribuições teóricas, agrupando teorias de outras áreas do conhecimento fundamentais para o desenvolvimento de agentes com consciência artificial, clarifica a arquitetura Baars-Franklin e realiza um estudo de caso da aplicação desse modelo na construção de um agente autônomo. / Abstract: For years, the human consciousness has challenged scientists from several areas of knowledge. In computer science, over the last decade, a significant growth in the number of studies about artificial consciousness has been observed. The architecture Baars-Franklin, developed by Stan Franklin, is a computational promising approach which is inspired by the Bernard Baars' global workspace theory. This work aims to discuss this technology and verify its advantages and disadvantages. Due to the intrinsic multidisciplinary characteristic of the study of artificial consciousness, first of all, the main general theories about consciousness are presented. After that, the most relevant studies about artificial consciousness are surveyed. Finally, the architecture Baars-Franklin is applied to control a virtual creature in an autonomous navigation problem. This dissertation brings some theoretical contributions, aggregating theories from other areas of knowledge, clarifying the architecture Baars-Franklin and showing a practical case study of the application of this model in order to build an autonomous agent. / Mestrado / Automação / Mestre em Engenharia Elétrica

Indiana University-Purdue University Indianapolis (IUPUI) / Multi-agent systems consist of multiple agents that interact and coordinate with each other to work towards to certain goal. Multi-agent systems naturally arise in a variety of domains such as robotics, telecommunications, and economics. The dynamic and complex nature of these systems entails the agents to learn the optimal solutions on their own instead of following a pre-programmed strategy. Reinforcement learning provides a framework in which agents learn optimal behavior based on the response obtained from the environment. In this thesis, we propose various novel de- centralized, learning automaton based algorithms which can be employed by a group of interacting learning automata. We propose a completely decentralized version of the estimator algorithm. As compared to the completely centralized versions proposed before, this completely decentralized version proves to be a great improvement in terms of space complexity and convergence speed. The decentralized learning algorithm was applied; for the first time; to the domains of distributed object tracking and distributed watershed management. The results obtained by these experiments show the usefulness of the decentralized estimator algorithms to solve complex optimization problems. Taking inspiration from the completely decentralized learning algorithm, we propose the novel concept of partial decentralization. The partial decentralization bridges the gap between the completely decentralized and completely centralized algorithms and thus forms a comprehensive and continuous spectrum of multi-agent algorithms for the learning automata. To demonstrate the applicability of the partial decentralization, we employ a partially decentralized team of learning automata to control multi-agent Markov chains. More flexibility, expressiveness and flavor can be added to the partially decentralized framework by allowing different decentralized modules to engage in different types of games. We propose the novel framework of heterogeneous games of learning automata which allows the learning automata to engage in disparate games under the same formalism. We propose an algorithm to control the dynamic zero-sum games using heterogeneous games of learning automata.

Thesis (M. Tech.) -- Central University of Technology, Free State, 2010 / Traditional corrective maintenance is both costly and ineffective. In some situations it is more cost effective to replace a device than to maintain it; however it is far more likely that the cost of the device far outweighs the cost of performing routine maintenance. These device related costs coupled with the profit loss due to reduced production levels, makes this reactive maintenance approach unacceptably inefficient in many situations. Blind predictive maintenance without considering the actual physical state of the hardware is an improvement, but is still far from ideal. Simply maintaining devices on a schedule without taking into account the operational hours and workload can be a costly mistake. The inefficiencies associated with these approaches have contributed to the development of proactive maintenance strategies. These approaches take the device health state into account. For this reason, proactive maintenance strategies are inherently more efficient compared to the aforementioned traditional approaches. Predicting the health degradation of devices allows for easier anticipation of the required maintenance resources and costs. Maintenance can also be scheduled to accommodate production needs. This work represents the design and simulation of an intelligent maintenance management system that incorporates device health prognosis with maintenance schedule generation. The simulation scenario provided prognostic data to be used to schedule devices for maintenance. A production rule engine was provided with a feasible starting schedule. This schedule was then improved and the process was determined by adhering to a set of criteria. Benchmarks were conducted to show the benefit of optimising the starting schedule and the results were presented as proof. Improving on existing maintenance approaches will result in several benefits for an organisation. Eliminating the need to address unexpected failures or perform maintenance prematurely will ensure that the relevant resources are available when they are required. This will in turn reduce the expenditure related to wasted maintenance resources without compromising the health of devices or systems in the organisation.

The ability to change one's beliefs in a rational manner is one of many facets of the abilities of an intelligent agent. Central to any investigation of belief change is the notion of an epistemic state. This dissertation is mainly concerned with three issues involving epistemic states: 1. How should an epistemic state be represented? 2. How does an agent use an epistemic state to perform belief change? 3. How does an agent arrive at a particular epistemic state? With regard to the first question, note that there are many different methods for constructing belief change operations. We argue that semantic constructions involving ordered pairs, each consisting of a set of beliefs and an ordering on the set of "possible worlds" (or equivalently, on the set of basic independent bits of information) are, in an important sense, more fundamental. Our answer to the second question provides indirect support for the use of semantic structures. We show how well-known belief change operations and related structures can be modelled semantically. Furthermore, we introduce new forms of belief change related operations and structures which are all defined, and motivated, in terms of such semantic representational formalisms. These include a framework for unifying belief revision and nonmonotonic reasoning, new versions of entrenchment orderings on beliefs, novel approaches to withdrawal operations, and an expanded view of iterated belief change. The third question is. one which has not received much attention in the belief change literature. We propose to extract extra-logical information from the formal representation of an agent's set of beliefs, which can then be used in the construction of epistemic state. his proposal is just a first approximation, although it seems to have the potential for developing into a full-fledged theory. / Computing / D.Phil.(Computer Science)