This thesis presents the design, development, and testing of a multi-agent systems laboratory that will enable the experimental investigation of Networked Control Systems. Networked Control Systems (NCS) are integrations of computation, networking, and physical dynamics, in which embedded devices are networked to sense, monitor, execute collaborative tasks, and interact with the physical world. As the potential for applications of NCS has increased, so has the research interest in this area. Possible applications include search and rescue, scientific data collection, and health care monitoring systems. One of the primary challenges in applying NCS is designing distributed algorithms that will enable the networked devices to achieve global objectives. Another challenge is in ensuring that distributed algorithms have the necessary robustness to achieve those global objectives in dynamic and unpredictable environments. A multi-agent systems laboratory provides the researcher with a means to observe the behavior and performance of distributed algorithms as they are executed on a set of networked devices. Through this observation, the researcher may discover robustness issues that were not present in computer simulation. The objective of this research is to design and implement the infrastructure for a multi-agent systems laboratory to observe distributed algorithms implemented on networked devices.

Le domaine scientifique dans lequel nos recherches s'insèrent est celui des Systèmes Multi-Agents. Dans ce domaine, nous nous intéressons à l'élaboration de modèles globaux définissant et structurant l'activité commune d'agents autonomes. En effet l'autonomie d'un agent se traduit par sa capacité à déterminer ses propres buts face à la configuration de l'environnement ainsi qu'à ses motivations. Du fait de cette autonomie, des conflits liés à l'existence de buts propres à chacun des agents autonomes d'une société d'agents peuvent être incompatibles, complémentaires ou en compétition avec les objectifs de la société. Des modèles de gestion de conflits existent, et dans le cadre de ce travail, nous nous sommes intéressés aux modèles organisationnels. Nous défendons la thèse selon laquelle la définition et la mise en place de contraintes globales influençant le fonctionnement d'agents autonomes doit être faite de manière à prendre en compte l'éventualité que les agents puissent ne pas respecter les contraintes selon les contextes et les objectifs individuels de chacun. Afin d'atteindre ces objectifs, nous proposons un modèle d'Institution Electronique comportant une description explicite et déclarative des contraintes à appliquer aux agents ainsi qu'un système de supervision et de renforcement de ces contraintes. Ce modèle permet de définir et de structurer grâce à MoiseInst l'activité d'un ensemble d'agents autonomes au sein d'une organisation normée et de superviser son respect grâce au système d'arbitrage Synai. Ce dernier est composé d'un ensemble d'agents de supervision capables de mettre en place des mécanismes de régulation et de renforcement de ces contraintes sur les agents autonomes. L'application du modèle se fait dans deux domaines totalement différents puisqu'ils concernent la création et l'exécution de contenu pour la télévision interactive d'un côté et la gestion de contrats électroniques d'une plate-forme de e-commerce de l'autre. Ils exhibent cependant la même problématique d'arbitrage d'entités autonomes devant respecter des contraintes. De plus, ces deux applications sont complémentaires par leur différence de portée de l'Institution.

This thesis addresses the problem of learning in physical heterogeneous multi-agent systems(MAS) and the analysis of the benefits of using heterogeneous MAS with respect tohomogeneous ones. An algorithm is developed for this task; building on a previous work on stability in distributed systems by Tad Hogg and Bernardo Huberman, and combining two phenomena observed in natural systems, task partition and hierarchical dominance. This algorithm is devised for allowing agents to learn which are the best tasks to perform on the basis of each agent's skills and the contribution to the team global performance. Agents learn by interacting with the environment and other teammates, and get rewards from the result of the actions they perform. This algorithm is specially designed for problems where all robots have to co-operate and work simultaneously towards the same goal. One example of such a problem is role distribution in a team of heterogeneous robots that form a soccer team, where all members take decisions and co-operate simultaneously. Soccer offers the possibility of conducting research in MAS, where co-operation plays a very important role in a dynamical and changing environment. For these reasons and the experience of the University of Girona in this domain, soccer has been selected as the test-bed for this research. In the case of soccer, tasks are grouped by means of roles.One of the most interesting features of this algorithm is that it endows MAS with a highadaptability to changes in the environment. It allows the team to perform their tasks, whileadapting to the environment. This is studied in several cases, for changes in the environment and in the robot's body. Other features are also analysed, especially a parameter that defines the fitness (biological concept) of each agent in the system, which contributes to performance and team adaptability.The algorithm is applied later to allow agents to learn in teams of homogeneous andheterogeneous robots which roles they have to select, in order to maximise team performance. The teams are compared and the performance is evaluated in the games against three hand-coded teams and against the different homogeneous and heterogeneous teams built in this thesis. This section focuses on the analysis of performance and task partition, in order to study the benefits of heterogeneity in physical MAS.In order to study heterogeneity from a rigorous point of view, a diversity measure is developed building on the hierarchic social entropy defined by Tucker Balch. This is adapted to quantify physical diversity in robot teams. This tool presents very interesting features, as it can be used in the future to design heterogeneous teams on the basis of the knowledge on other teams.

Truck transportation is considered as a favourable mode by shippers to carry freight at most ranges of distance as it has more flexibility in fleet size, capacity, scheduling, routing, and access. Although truck is considered as the popular mode for freight transportation, road-rail intermodal freight transportation becomes an attractive alternative to road only mode since the latter has no longer assured a reliable service due to traffic congestion problem. It also raises public concern in environmental and road safety impacts. Intermodal freight transportation is defined as a system that carries freight from origin to destination using two or more transportation modes where transfers between modes occur at an intermodal freight terminal. Success of the terminal depends on four major factors, namely: location, efficiency, financial sustainability, and rail level of service. Among these, the location is one of the most crucial success factors and needs to be considered carefully as it has direct and indirect impacts on a number of stakeholders including terminal users, terminal operators, transport network infrastructure providers, and community. Limitations of previous terminal location evaluation models in representing individual preference and behaviour as well as accommodating negotiation and communication between the players bring in an opportunity to develop a new model which is more flexible and capable of providing a solution that is not necessary to be optimal, but acceptable for every player without requiring explicit trade-offs. This thesis is aimed at demonstrating the feasibility of applying an agent-based approach to the evaluation of intermodal freight terminal location and investigating terminal effectiveness against stakeholder equity and some important aspects arising from the different stakeholders’ viewpoints. Agent technologies were introduced to model the stakeholders as individual agents. The agent concept was adopted to develop a decentralised location evaluation system that is able to balance the terminal effectiveness with the stakeholder equity. The proposed agent-based location evaluation model was modelled as a hierarchical control system that comprises three decision levels: local level, stakeholder level and policy level. Policy level is the highest decision level, which is represented by a policy maker. Apart from the policy level, the rest can be viewed as operational decision levels. Local level is the lowest control level. At this level, each stakeholder was classified into stakeholder groups based on their characteristics and interest. The terminal scenarios were then evaluated based on benefit maximisation criteria. Stakeholder control is the higher control level than the local level. It represents the control level where negotiations and decisions between groups of people (stakeholders) with different point of views are made. At this level, negotiation process was used to determine terminal location based on preference and equity of stakeholders. The determined terminal site was then used in the evaluation against constraints to ensure that all agents are satisfied. The terminal location decision for South East Queensland (SEQ) was applied as a case study of this thesis. The SEQ strategic freight transport model was developed, calibrated, and validated to assist in providing inputs for the evaluation of terminal location. The results indicated that for the developed agent-based location evaluation model, Yatala was selected as the most appropriate terminal location that results in the highest effectiveness and equity (as measured by level of satisfaction and Gini coefficient, respectively). Other location evaluation models were also used in comparison with the developed agent-based location evaluation model. Those include P-Median, P-Centre, and maximum covering models. It was found that the agent-based location evaluation model outperformed the other location evaluation models. Finally, a sensitivity analysis was conducted in order to evaluate the consistency of model outputs against the uncertainties in the input parameters. In most cases, the terminal location decisions obtained from the developed agent-based location evaluation model was not sensitive to the changes in those parameters. However, the results suggested that when a unit cost of truck travel delay increased, the impact on the final terminal location decisions was observed. This thesis demonstrated the feasibility of applying a decentralised approach to terminal location decision problem using a multi-agent concept and evaluating it against other well-known location problems. A new framework and methodology for the planning of intermodal terminal location evaluation was also formulated. Finally, the problems of terminal location evaluation and optimisation of intermodal freight terminal operation were integrated into a single evaluation model.

Engineering complex systems and New Product Development (NPD) are major challenges for contemporary engineering design and must be studied at three levels of: Products, Processes and Organizations (PPO). The science of complexity indicates that complex systems share a common characteristic: they are robust yet fragile. Complex and large scale systems are robust in the face of many uncertainties and variations; however, they can collapse, when facing certain conditions. This is so since complex systems embody many subtle, intricate and nonlinear interactions. If formal modelling exercises with available computational approaches are not able to assist designers to arrive at accurate predictions, then how can we immunize our large scale and complex systems against sudden catastrophic collapse? This thesis is an investigation into complex product design. We tackle the issue first by introducing a template and/or design methodology for complex product design. This template is an integrated product design scheme which embodies and combines elements of both design theory and organization theory; in particular distributed (spatial and temporal) problem solving and adaptive team formation are brought together. This design methodology harnesses emergence and innovation through the incorporation of massive amount of numerical simulations which determines the problem structure as well as the solution space characteristics. Within the context of this design methodology three design methods based on measures of complexity are presented. Complexity measures generally reflect holistic structural characteristics of systems. At the levels of PPO, correspondingly, the Immunity Index (global modal robustness) as an objective function for solutions, the real complexity of decompositions, and the cognitive complexity of a design system are introduced These three measures are helpful in immunizing the complex PPO from chaos and catastrophic failure. In the end, a conceptual decision support system (DSS) for complex NPD based on the presented design template and the complexity measures is introduced. This support system (IMMUNE) is represented by a Multi Agent Blackboard System, and has the dual characteristic of the distributed problem solving environments and yet reflecting the centralized viewpoint to process monitoring. In other words IMMUNE advocates autonomous problem solving (design) agents that is the necessary attribute of innovative design organizations and/or innovation networks; and at the same time it promotes coherence in the design system that is usually seen in centralized systems.

The thesis solves a challenging problem in military logistics for tasks such as transportation scheduling and combinatorial auctions. A conceptual model has been developed that captures the organisational business processes involved and an effective implementation suitable for computer software agents. The protocol facilitates planning and task allocation among organisations in decentralised, dynamic and open environments.

Η παρούσα εργασία πραγματεύεται ζητήματα που αφορούν σε συστήματα αυτομάτου ελέγχου με ιδιαίτερη έμφαση στα συστήματα κατανεμημένου ελέγχου. Σε αυτό το πλαίσιο, πραγματοποιείται στην εργασία καταρχάς μια σύντομη ανασκόπηση σε σχέση με τα συστήματα ελέγχου και με το πώς αυτά τα συστήματα εισήχθησαν για χρήση στη βιομηχανία. Στη συνέχεια, παρουσιάζονται τα κατανεμημένα συστήματα ελέγχου, τα βασικά χαρακτηριστικά τους αλλά και οι βασικές μεθοδολογίες σχεδίασης ενός τέτοιου συστήματος. Γίνεται επίσης σημαντική αναφορά στα πολυπρακτορικά συστήματα (Multi Agent Systems – MAS) τα οποία αποτελούν το μέλλον της ανάπτυξης των κατανεμημένων συστημάτων. Τέλος, συνοψίζονται τα βασικά προβλήματα, οι τάσεις αλλά και η κατάσταση στην αγορά αναφορικά με αυτά τα συστήματα. / -

abstract: TaxiWorld is a Matlab simulation of a city with a fleet of taxis which operate within it, with the goal of transporting passengers to their destinations. The size of the city, as well as the number of available taxis and the frequency and general locations of fare appearances can all be set on a scenario-by-scenario basis. The taxis must attempt to service the fares as quickly as possible, by picking each one up and carrying it to its drop-off location. The TaxiWorld scenario is formally modeled using both Decentralized Partially-Observable Markov Decision Processes (Dec-POMDPs) and Multi-agent Markov Decision Processes (MMDPs). The purpose of developing formal models is to learn how to build and use formal Markov models, such as can be given to planners to solve for optimal policies in problem domains. However, finding optimal solutions for Dec-POMDPs is NEXP-Complete, so an empirical algorithm was also developed as an improvement to the method already in use on the simulator, and the methods were compared in identical scenarios to determine which is more effective. The empirical method is of course not optimal - rather, it attempts to simply account for some of the most important factors to achieve an acceptable level of effectiveness while still retaining a reasonable level of computational complexity for online solving. / Dissertation/Thesis / M.S. Computer Science 2011

Dentro da Educação a Distância (EaD), os softwares de apoio como os Ambientes Virtuais de Aprendizagem (AVA) são considerados recursos que favorecem a comunicação entre os atores envolvidos, permitindo a troca de informação. Atribuir Inteligência Artificial a esses AVAs, utilizando Sistemas Multiagentes (SMA) e uma forma de procurar que os mesmos tenham um bom desempenho e que seus recursos facilitem o processo de aprendizagem. Esse trabalho cont em um estudo sobre os principais AVAs existentes e sobre os métodos alternativos de integração de AVA com SMA. Ao analisar o estado da arte dos AVAs pode-se observar que todos trabalham como ferramentas de auxílio ao aluno, por em nenhum deles trabalha aspectos da gestão da EaD dando suporte aos aspectos relevantes da legislação dessa modalidade. Por esse motivo, essa Tese tem por objetivo a criação de um modelo de integração AVA-SMA que possa tornar o AVA MOODLE capaz de auxiliar os gestores da EaD em suas diferentes tarefas, com base na incorporação, ao mesmo, de um modelo de representação de legislação. Para realizar essa integração do modelo AVA-SMA orientado a Legislação foi desenvolvido um espec co modelo organizacional de Sistema Multiagente. Por m com base em um estudo de caso, ser a realizado simulações para veri car as funcionalidades do Modelo de Sistema AVA-SMA orientado a Legislação, proposto nesta Tese. / In Distance Learning (EaD), supporting software such as Virtual Learning Environments (VLE) are considered resources that favor communication between the actors involved, allowing the exchange of information. Assigning Arti cial Intelligence to these VLEs, using Multi-Agent Systems (MAS) is a way of ensuring they have a good performance and that its resources facilitate the learning process. This work contains a study on the major existing VLEs and on alternative methods to integrate VLE with MAS. When analyzing the state of the art of the VLEs it is possible to see that all of them work as aid tools for students, but none of them work on management aspects of distance learning that support the relevant aspects of the legislation for this type of education. Therefore, this thesis aims to create a VLE-MAS integration model that can make the VLE MOODLE able to help distance learning managers in their di erent tasks, based on incorporating a legislation representation model to it. To accomplish this integration of the legislation-oriented VLE-MAS model, a speci c Multi-Agent System organizational model was developed. At last, based on a case study, simulations will be conducted to verify the functionalities of the VLE-MAS System Model oriented to legislation, proposed in this thesis.

Este trabalho apresenta o projeto e arquitetura de uma plataforma para execução de Agentes com funções implementadas em hardware, tomando vantagem do uso de hardware reconfigurável. Os Agentes em hardware são implementados utilizando dispositivos FPGA (Field-programmable Gate Array). O trabalho estende trabalhos anteriores semelhantes na área, com o diferencial de adicionar suporte às funcionalidades de reconfiguração parcial do hardware, suportar aplicações que demandam alto desempenho em hardware – como processamento de sinais e imagens – e redução de recursos de hardware necessários para execução da interface em software. A plataforma proposta utiliza o framework JADE (Java Agent Development Framework), que é um dos frameworks mais populares no estado da arte de desenvolvimento de Agentes e compatível com outros frameworks de Agentes através da conformidade aos padrões FIPA (Foundation for Intelligent Physical Agents). Com o uso do JADE, a plataforma possibilita a comunicação entre Agentes com funções implementadas em hardware e Agentes puramente implementados em software dentro de um mesmo SMA (Sistema Multi-Agente). Uma funcionalidade notável do JADE é a possibilidade de migração de Agentes entre plataformas de um mesmo SMA. Através do uso da reconfiguração parcial de hardware em conjunto com o JADE, a plataforma permite a migração de Agentes de software para hardware e vice-versa, além de suportar reconfiguração de múltiplos Agentes em hardware com um único FPGA. A plataforma faz uso de um único chip através do uso de um processador soft core implementado na lógica programável. O uso deste processador é um diferencial neste trabalho, e mostra que é possível utilizar o JADE em sistemas embarcados com recursos de processamento limitados. Ou seja, em um Agente cuja principal função é implementada em hardware, basta um processador bastante simples para atuar como uma interface entre o hardware e o framework de Agentes. O uso do processador dentro do FPGA tem também o benefício de oferecer formas de acesso mais integrado ao hardware, permitindo maior desempenho na transmissão de dados ao hardware. A plataforma foi validada através de estudos de caso de Agentes com implementações em hardware e em software, incluindo um estudo de caso aplicado de processamento de imagem embarcado utilizando VANTs (Veículos Aéreos Não-Tripulados). O estudo também apresenta comparações de desempenho entre a execução dos Agentes em hardware e em outras plataformas embarcadas de prateleira. Os experimentos realizados mostram um ganho significativo de desempenho nas implementações em FPGA, especialmente considerando processamento de imagens de alta resolução, mesmo considerando que o FPGA executa em frequências consideravelmente reduzidas em comparação às outras plataformas testadas. / This work described the design and architecture of a platform for execution of Agents whose functions are implemented in hardware, by leveraging the use of reconfigurable hardware. The hardware Agents are implemented using FPGA (Field-programmable Gate Array) devices. This work extends previous similar work in this field, while adding the features of hardware partial reconfiguration, supporting applications which require high performance in hardware – such as image or signal processing – and reducing the hardware resource for the software interface execution. The proposed platform makes use of the JADE (Java Agent Development Framework) framework, which is one of the most popular frameworks in state-of-the-art Agent development, and is also compatible with other Agent development frameworks due to compliance with FIPA (Foundation for Intelligent Physical Agents) standards. With the use of JADE, the platform enables communication among Agents which are implemented in hardware and Agents purely implemented in software, inside the same MAS (Multi-Agent System). One notable feature of JADE is the possibility of migrating Agents among platforms inside a single MAS. Through the use of hardware partial reconfiguration along with JADE, the platform enables the migration of Agents from software to hardware and viceversa, in addition to supporting múltiple hardware Agents in a single FPGA. The platform makes use of a single chip, by using a MicroBlaze soft core processor implemented in programmable logic. The use of this processor is a distinction on this work, and it shows that it is possible to use JADE on embedded systems with limited processing power. That is, in an Agent whose main function is implemented in hardware, a very simple processor to act as an interface between hardware and the Agent framework is enough. The use of the soft core processor inside the FPGA also has the benefit of offering more integrated ways of accessing hardware, enabling higher performance for transferring data to hardware. The platform was validated through case studies of hardware and software Agent implementation, including a case study applied to image processing using UAVs (Unmanned Aerial Vehicles). The study also shows performance comparisons between the Agent execution in hardware and in other off-the-shelf embedded platforms. The performed experiments report a significative performance increase in the FPGA implementations, particularly in high resolution image processing, even considering that the FPGA runs in considerably lower clock frequency than the other tested platforms.