Agrupamento de dados é o processo que consiste em dividir um conjunto de dados em grupos de forma que dados semelhantes entre si permaneçam no mesmo grupo enquanto que dados dissimilares sejam alocados em grupos diferentes. Técnicas tradicionais de agrupamento de dados têm sido usualmente desenvolvidas de maneira centralizada dependendo assim de estruturas que devem ser acessadas e modificadas a cada passo do processo de agrupamento. Além disso, os resultados gerados por tais métodos são dependentes de informações que devem ser fornecidas a priori como por exemplo número de grupos, tamanho do grupo ou densidade mínima/máxima permitida para o grupo. O presente trabalho visa propor o bee clustering, um algoritmo distribuído inspirado principalmente em técnicas de inteligência de enxames como organização de colônias de abelhas e alocação de tarefas em insetos sociais, desenvolvido com o objetivo de resolver o problema de agrupamento de dados sem a necessidade de pistas sobre o resultado desejado ou inicialização de parâmetros complexos. O bee clustering é capaz de formar grupos de agentes de maneira distribuída, uma necessidade típica em cenários de sistemas multiagente que exijam capacidade de auto-organização sem controle centralizado. Os resultados obtidos mostram que é possível atingir resultados comparáveis as abordagens centralizadas. / Clustering can be defined as a set of techniques that separate a data set into groups of similar objects. Data items within the same group are more similar than objects of different groups. Traditional clustering methods have been usually developed in a centralized fashion. One reason for this is that this form of clustering relies on data structures that must be accessed and modified at each step of the clustering process. Another issue with classical clustering methods is that they need some hints about the target clustering. These hints include for example the number of clusters, the expected cluster size, or the minimum density of clusters. In this work we propose a clustering algorithm that is inspired by swarm intelligence techniques such as the organization of bee colonies and task allocation among social insects. Our proposed algorithm is developed in a decentralized fashion without any initial information about number of classes, number of partitions, and size of partition, and without the need of complex parameters. The bee clustering algorithm is able to form groups of agents in a distributed way, a typical necessity in multiagent scenarios that require self-organization without central control. The performance of our algorithm shows that it is possible to achieve results that are comparable to those from centralized approaches.

Agrupamento de dados é o processo que consiste em dividir um conjunto de dados em grupos de forma que dados semelhantes entre si permaneçam no mesmo grupo enquanto que dados dissimilares sejam alocados em grupos diferentes. Técnicas tradicionais de agrupamento de dados têm sido usualmente desenvolvidas de maneira centralizada dependendo assim de estruturas que devem ser acessadas e modificadas a cada passo do processo de agrupamento. Além disso, os resultados gerados por tais métodos são dependentes de informações que devem ser fornecidas a priori como por exemplo número de grupos, tamanho do grupo ou densidade mínima/máxima permitida para o grupo. O presente trabalho visa propor o bee clustering, um algoritmo distribuído inspirado principalmente em técnicas de inteligência de enxames como organização de colônias de abelhas e alocação de tarefas em insetos sociais, desenvolvido com o objetivo de resolver o problema de agrupamento de dados sem a necessidade de pistas sobre o resultado desejado ou inicialização de parâmetros complexos. O bee clustering é capaz de formar grupos de agentes de maneira distribuída, uma necessidade típica em cenários de sistemas multiagente que exijam capacidade de auto-organização sem controle centralizado. Os resultados obtidos mostram que é possível atingir resultados comparáveis as abordagens centralizadas. / Clustering can be defined as a set of techniques that separate a data set into groups of similar objects. Data items within the same group are more similar than objects of different groups. Traditional clustering methods have been usually developed in a centralized fashion. One reason for this is that this form of clustering relies on data structures that must be accessed and modified at each step of the clustering process. Another issue with classical clustering methods is that they need some hints about the target clustering. These hints include for example the number of clusters, the expected cluster size, or the minimum density of clusters. In this work we propose a clustering algorithm that is inspired by swarm intelligence techniques such as the organization of bee colonies and task allocation among social insects. Our proposed algorithm is developed in a decentralized fashion without any initial information about number of classes, number of partitions, and size of partition, and without the need of complex parameters. The bee clustering algorithm is able to form groups of agents in a distributed way, a typical necessity in multiagent scenarios that require self-organization without central control. The performance of our algorithm shows that it is possible to achieve results that are comparable to those from centralized approaches.

Orientadores: Ricardo Ribeiro Gudwin, Leandro Nunes de Castro Silva / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-11T06:40:01Z (GMT). No. of bitstreams: 1 Pereira_AndreLuizVizine_M.pdf: 1817378 bytes, checksum: 1d28283d8d2855800dd0f406eb97e5e0 (MD5) Previous issue date: 2007 / Resumo: A Internet tornou-se um dos principais meios de comunicação da atualidade, reduzindo custos, disponibilizando recursos e informação para pessoas das mais diversas áreas e interesses. Esta dissertação desenvolve e aplica duas abordagens de computação inspirada na biologia aos problemas de otimização do processo de busca e recuperação de informação na web e agrupamento de textos. Os algoritmos investigados e modificados são o algoritmo genético e o algoritmo de agrupamento por colônia de formigas. O objetivo final do trabalho é desenvolver parte do conjunto de ferramentas que será usado para compor o núcleo de uma comunidade virtual acadêmica adaptativa. Os resultados obtidos mostraram que o algoritmo genético é uma ferramenta adequada para otimizar a busca de informação na web, mas o algoritmo de agrupamento por colônia de formigas ainda apresenta limitações quanto a sua aplicabilidade para agrupamento de textos. / Abstract: The Internet became one of the main sources of information and means of communication, reducing costs and providing resources and information to the people all over the world. This dissertation develops and applies two biologically-inspired computing approaches, namely a genetic algorithm and the ant-clustering algorithm, to the problems of optimizing the information search and retrieval over the web, and to perform text clustering. The final goal of this project is to design and develop some of the tools to be used to construct an adaptive academic virtual community. The results obtained showed that the genetic algorithm can be feasibly applied to the optimizing information search and retrieval, whilest the ant-clustering algorithm needs further investigation in order to be efficiently applied to text clustering. / Mestrado / Engenharia de Computação / Mestre em Engenharia Elétrica

A Wireless Sensor Network (WSN) is a group of spatially dispersed sensors that monitor the physical conditions of the environment and collect data at a central location. Sensor deployment is one of the main design aspects of WSNs as this a ffects network coverage. In general, WSN deployment methods fall into two categories: planned deployment and random deployment. This thesis considers planned sensor deployment of a Mobile Wireless Sensor Network (MWSN), which is defined as selectively deciding the locations of the mobile sensors under the given constraints to optimize the coverage of the network. Metaheuristic algorithms are powerful tools for the modeling and optimization of problems. The Krill Herd Algorithm (KHA) is a new nature-inspired metaheuristic algorithm which can be used to solve the sensor deployment problem. A Multi-Agent System (MAS) is a system that contains multiple interacting agents. These agents are autonomous entities that interact with their environment and direct their activity towards achieving speci c goals. Agents can also learn or use their knowledge to accomplish a mission. Multi-agent systems can solve problems that are very difficult or even impossible for monolithic systems to solve. In this work, a modification of KHA is proposed which incorporates MAS to obtain a Multi-Agent Krill Herd Algorithm (MA-KHA). To test the performance of the proposed method, five benchmark global optimization problems are used. Numerical results are presented which show that MA-KHA performs better than the KHA by finding better solutions. The proposed MA-KHA is also employed to solve the sensor deployment problem. Simulation results are presented which indicate that the agent-agent interactions in MA-KHA improves the WSN coverage in comparison with Particle Swarm Optimization (PSO), the Firefly Algorithm (FA), and the KHA. / Graduate

In my doctoral dissertation, I tackled two of the main open problems in swarm robotics: design and verification. I did so by using model checking.<p>Designing and developing individual-level behaviors to obtain a desired swarm-level goal is, in general, very difficult, as it is difficult to predict and thus design the non-linear interactions of tens or hundreds individual robots that result in the desired collective behavior. In my dissertation, I presented my novel contribution to the top-down design of robot swarms: property-driven design. Property-driven design is based on prescriptive modeling and model checking. Using property-driven design it is possible to design robot swarms in a systematic way, realizing systems that are "correct by design". I demonstrated property-driven design on two case-studies: aggregation and foraging.<p>Developing techniques to analyze and verify a robot swarm is also a necessary step in order to employ swarm robotics in real-world applications. In my dissertation, I explored the use of model checking to analyze and verify the properties of robot swarms. Model checking allows us to formally describe a set of desired properties of a system, in a more powerful and precise way compared to other mathematical approaches, and verify whether a given model of a system satisfies them. I explored two different approaches: the first based on Bio-PEPA and the second based on KLAIM. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

In this dissertation, we propose and study methods for information transfer within a swarm of mobile robots that coordinately move, or flock, in a common direction. We define information transfer as the process whereby robots share directional information in order to coordinate their heading direction. We identify two paradigms of information transfer: explicit information transfer and implicit information transfer. <p><p>In explicit information transfer, directional information is transferred via communication. Explicit information transfer requires mobile robots equipped with a a communication device. We propose novel communication strategies for explicit information transfer, and we perform flocking experiments in different situations: with one or two desired directions of motion that can be static or change over time. We perform experiments in simulation and with real robots. Furthermore, we show that the same explicit information transfer strategies can also be applied to another collective behavior: collective transport with obstacle avoidance. <p><p>In implicit information transfer, directional information is transferred without communication. We show that a simple motion control method is sufficient to guarantee cohesive and aligned motion without resorting to communication or elaborate<p>sensing. We analyze the motion control method for its capability to achieve flocking with and without a desired direction of motion, both in simulation and using real robots. Furthermore, to better understand its underlying mechanism, we study this<p>method using tools of statistical physics, showing that the process can be explained in terms of non-linear elasticity and energy-cascading dynamics. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

A visão de muitas pessoas sobre uma colônia de formigas, em geral, é de que estes pequenos e inofensivos insetos somente se movem aleatoriamente para coletar alimento e conservá-los em seus ninhos. Um olhar destreinado não conseguiria notar o nível de complexidade e organização que é requerido por uma colônia de formigas para sua sobrevivência. Uma formiga simples é parte de um grande grupo que coopera entre si para criar um superorganismo. Sem uma autoridade central ou indivíduos com habilidade de um pensamento cognitivo complexo, a colônia se auto-organiza, e, de fato, ajusta seus recursos de uma maneira muito eficiente. Essa dissertação investiga o papel da comunicação indireta nas tarefas de exploração e forrageamento, e como isso afeta as decisões de um agente simples e traz um comportamento emergente útil à toda colônia. Por fim, este trabalho implementa uma plataforma de simulação multi-agente inspirado em formigas / Most people\'s view of an ant colony and ants in general is that they simply pose harmless little insects that move randomly and gather food in their underground nests. The untrained eye would have never guessed the level of complexity and organisation that is required in order for an ant colony to survive. The simple ant is a part of a huge group, which cooperate one superorganism. Without any central authority or the ability of complex cognitive thought from the individuals, the colony seems to self organise and in fact adjust its resources in a quite efficient way. This dissertation investigates the role of indirect communication in the exploration and forage task and how it affects the decisions of the single agent and brings an emergent behaviour that is useful to all the colony. Finally this work implements an ant inspired multi-agent simulation plataform

Das Forschungsfeld Schwarmintelligenz, also die Anwendung des Verhaltens dezentraler selbstorganisierender Tierkollektive, im Kontext der Informatik hat eine Reihe von state-of-the-art Kontroll- und Optimierungsmechanismen hervorgebracht. Die Untersuchung selbstorganisierender biologischer Systeme fördert zum einen das Design neuer robuster und adaptiver Algorithmen. Zum anderen kann sie das Verständnis der Funktionalität von selbstorganisierenden Prinzipien, welche in der Natur auftreten, unterstützen. Diese Arbeit deckt beide zuvor beschriebenen Aspekte ab. Unter Verwendung von Modellen und Simulation werden offene Fragen bezüglich der Organisation und des Verhaltens von sozialen Insekten beleuchtet. Weiter werden Abstraktionen von selbstorganisierenden Konzepten, welche man bei sozialen Insekten findet, genutzt, um neue Methoden zur Optimierung zu entwickeln. Der erste Teil dieser Arbeit untersucht allgemeine Aspekte der Arbeitsteilung sozialer Insekten. Zuerst wird die Anpassungsfähigkeit von unterschiedlich großen Kolonien, bezüglich dynamischer Veränderungen in der Umwelt untersucht. Die Ergebnisse zeigen, dass die Fähigkeit einer Kolonie, auf Veränderung in der Umwelt zu reagieren, von der Koloniegröße beeinflusst wird. Ein weiterer Aspekt der Arbeitsteilung, welcher in dieser Arbeit untersucht wird, ist, inwieweit eine räumliche Verteilung von Aufgaben und Individuen einen Einfluss auf die Arbeitsteilung hat. Die Ergebnisse deuten an, dass soziale Insekten von einer räumlichen Trennung, der zu bewerkstelligenden Aufgaben profitieren, da eine solche Trennung die Produktivität der Kolonie erhöht. Das könnte erklären, warum eine räumliche getrennte Anordnung von Aufgaben und Individuen häufig in realen Kolonien sozialer Insekten beobachtet werden kann. Der zweite Teil dieser Arbeit untersucht verschiedene Aspekte von Selbstorganisation bei Honigbienen. Zunächst wird der Einfluss der räumlichen Verteilung von Nestplätzen auf die Nestplatzsuche der europäischen Honigbiene Apis mellifera untersucht. Die Ergebnisse legen nahe, dass die Nestplatzsuche eines Schwarms aktiv durch die Anordnung der Nestplätze in der Umwelt beeinflusst wird. Eine nestplatzreiche Umgebung kann den Prozess eines Schwarms, sich für einen Nestplatz zu entscheiden, stark behindern. Das könnte erklären, warum Honigbienenarten, die geringe Anforderungen an Nestplätze haben, was die Anzahl von potenziellen Nestplätzen natürlich erhöht, eine sehr ungenaue Form der Nestplatzsuche aufweisen. Ein zweiter Aspekt der Honigbienen, welcher untersucht wird, sind die Steuerungsmechanismen, die dem kollektiven Flug eines Bienenschwarms unterliegen. Zwei mögliche Führungsmechanismen, aktive und passive Führung, werden hinsichtlich ihrer Fähigkeit verglichen, die Flugeigenschaften eines echten Honigbienenschwarms zu reproduzieren. Die Simulationsergebnisse bestätigen aktuelle empirische Befunde und zeigen, dass aktive Führung in der Lage ist, Charakteristika fliegender Schwärme widerzuspiegeln. Bei passiver Führung ist das nicht der Fall. Eine Anwendung biologischer Konzepte im Bereich der Informatik wird anhand der Nestplatzsuche demonstriert. Diese ist ein natürlicher Optimierungsprozess, basierend auf einfachen Regeln. Erzielt wird eine lokale Optimierung, die es einem Schwarm ermöglicht, Nestplätze in einer bisher unbekannten Umgebung zu finden und aus diesen den besten Nestplatz zu wählen. Das ist die Motivation, Nestplatzsuche im Bereich der Optimierung anzuwenden. Hierfür wird zuerst das Optimierungspotenzial der biologischen Nestplatzsuche mit Hilfe eines biologischen Modells untersucht. Basierend auf der Nestplatzsuche wird ein abstrahiertes algorithmisches Schema, das so genannte „Bee Nest-Site Selection Scheme“ (BNSSS) entworfen. Basierend auf dem Schema wird der erste Nestplatzsuche inspirierte Optimierungsalgorithmus „Bee-Nest\\\''\\\'' für die Anwendung im Bereich von molekular Docking entwickelt. Im Vergleich zu anderen Optimierungsalgorithmen erzielt „Bee-Nest“ eine sehr gute Leistung. / The application in computer science of the behaviour found in decentralized self-organizing animal collectives -- also known as swarm intelligence -- has brought forward a number of state-of-the art control and optimization mechanisms. Further study of such self-organizing biological systems can foster the design of new robust and adaptive algorithms, as well as aid in the understanding of self-organizing processes found in nature. This thesis covers both of the aspects described above, namely the use of computational models to investigate open questions regarding the organization and behaviour of social insects, as well as using the abstraction of concepts found in social insects to generate new optimization methods. In the first part of this work, general aspects of division of labour in social insects are investigated. First the adaptiveness of different-sized colonies to dynamic changes in the environment is analysed. The findings show that a colony\\\''s ability to react to changes in the environment scales with its size. Another aspect of division of labour which is investigated is the extent to which different spatial distributions of tasks and individuals influence division of labour. The results suggest that social insects can benefit from a spatial separation of tasks within their environment, as this increases the colony\\\''s productivity. This could explain why a spatial organization of tasks and individuals is often observed in real social insect colonies. The second part of this work investigates several aspects of self-organization found in honeybees. First the influence of spatial nest-site distribution on the ability of the European honeybee Apis mellifera to select a new nest-site is studied. The results suggest that a swarm\\\''s habitat can influence its decision-making process. Nest-site rich habitats can obstruct a swarm\\\''s ability to choose a single site if all sites are of equal quality. This could explain why in nature honeybee species which have less requirements regarding a new nest-site have evolved a more imprecise form of nest-site selection than cavity-nesting species. Another aspect of honeybees which is investigated is the guidance behaviour in migrating swarms. Two potential guidance mechanisms, active and passive guidance, are compared regarding their ability to reproduce real honeybee swarm flight characteristics. The simulation results confirm previous empirical findings, as they show that active guidance is able to reflect a number of characteristics which can be observed in real moving honeybee swarms, while this is not the case for passive guidance. Nest-site selection in honeybees can be regarded as a natural optimization process. It is based on simple rules and achieves local optimization as it enables a swarm to decide between several potential nest-sites in a previously unknown dynamic environment. These factors motivate the application of the nest-site selection process to the problem domain of function optimization. First, the optimization potential of the biological nest-site selection process is studied. Then a general algorithmic scheme called ``Bee Nest-Site Selection Scheme\\\''\\\'' (BNSSS) is introduced. Based on the scheme the first nest-site inspired optimization algorithm ``Bee-Nest\\\''\\\'' is introduced and successfully applied to the domain of molecular docking.

The intention of the dissertation is the applied research of the collective ( group ) ( swarm ) intelligence . To demonstrate the applicability of the collective intelligence, the Particle Swarm Optimization ( PSO ) algorithm has been studied in which the problem of the collective intelligence is transferred to mathematical optimization in which the particle swarm searches for a global optimum within the defined problem space, and the searching is controlled according to the pre-defined objective function which represents the solved problem. A new search strategy has been designed and experimentally tested in which the particles continuously adjust their behaviour according to the characteristics of the problem space, and it has been experimentally discovered how the impact of the objective function representing a solved problem manifests itself in the behaviour of the particles. The results of the experiments with the proposed search strategy have been compared to the results of the experiments with the reference version of the PSO algorithm. Experiments have shown that the classical reference solution, where the only condition is a stable trajectory along which the particle moves in the problem space, and where the influence of a control objective function is ultimately eliminated, may fail, and that the dynamic stability of the trajectory of the particle itself is not an indicator of the searching ability nor the convergence of the algorithm to the true global solution of the solved problem. A search strategy solution has been proposed in which the PSO algorithm regulates its stability by continuous adjustment of the particles behaviour to the characteristics of the problem space. The proposed algorithm influenced the evolution of the searching of the problem space, so that the probability of the successful problem solution increased.

付記する学位プログラム名: 京都大学卓越大学院プログラム「先端光・電子デバイス創成学」 / 京都大学 / 新制・課程博士 / 博士(工学) / 甲第25290号 / 工博第5249号 / 新制||工||1999(附属図書館) / 京都大学大学院工学研究科電気工学専攻 / (主査)教授 阪本 卓也, 教授 引原 隆士, 准教授 薄 良彦, 教授 土居 伸二 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DFAM