L'objectif de cette thèse est d'étudier des propriétés des points d'équilibre dans des réseaux de transport multi-critères et de développer des méthodes numériques permettant de trouver l'ensemble des points d'équilibre ou une partie représentative de cet ensemble. Le travail est structure comme suit. Dans le premier chapitre nous donnons une introduction de la thèse. Le chapitre 2 est un rappel de certaines notions que nous utilisons dans les autres. Nous y rappelons le concept de point optimal de Pareto, les fonctions multivoques et les problèmes d'inégalité variationnelle. Nous introduisons certaines fonctions de scalarisation et puis établissons quelques propriétés importantes. Dans le chapitre 3, nous décrivons les réseaux de transport qui sont étudiés dans cette thèse. Dans chaque modèle, nous rappelons les définitions des points d'équilibre et donnons une relation entre ces définitions. Dans le chapitre 4 nous traitons les réseaux de transport multi-critères mono-produit sans contraintes de capacité. Tout d'abord, nous construisons deux problèmes d'optimisation dont les solutions sont exactement l'ensemble des points d'équilibre du modèle initiale et établissons certaines propriétés importantes de continuité et de dérivabilité génériques des fonctions objectifs. Puis nous donnons une formule permettant de calculer le gradient des fonctions objectifs. Nous proposons également un algorithme et prouvons sa convergence pour générer une représentation de l'ensemble des points d'équilibre. Puisque les fonctions objectifs de nos problèmes d'optimisation ne sont pas continues, une méthode de lissage est également considérée afin d'utiliser quelques techniques d'optimisation globale. En fin, nous introduisons le concept de point d'équilibre robuste, puis nous établissons des critères de robustesse et une formule permettant de calculer le rayon de robustesse. Dans le chapitre 5 nous étudions des points d'équilibre vectoriel dans le réseau de transport multi-critères mono-produit sous contraintes de capacité.Tout d'abord, nous proposons un problème d'optimisation équivalent. En utilisant des techniques analogues à celles du chapitre 4 nous obtenons également un sous-ensemble des points d'équilibre du modèle proposé. Dans le dernier chapitre nous considérons des points d'équilibre fort dans le réseau de transport multi-critères multi-produit sous contraintes de capacité. Nous établissons des conditions d'existence des points d'équilibre fort, des relations entre des points d'équilibre fort et des points d'équilibre par rapport à une famille de fonctions ainsi qu'une relation entre des points d'équilibre fort et les points efficaces de l'ensemble des valeurs de la fonction de coût. En plus nous construisons des problèmes d'inéqualité variationnellle, dont les solutions sont les points d'équilibre fort. La dernière partie de ce chapitre est consacrée à un algorithme permettant de trouver des points d'équilibre d'un réseau multi-critères sous contraintes de capacité. Certains exemples numériques sont donnés pour illustrer notre méthode. Nous fermons la thèse avec une liste de références et appendice contenant le code matlab de nos algorithmes. / The purpose of this thesis is to study equilibria in multi-criteria trafficnetworks and develop numerical methods to find the set of all equilibria oronly one representative part of this set. The thesis is structured as follows.In the first chapter we present an introduction of the thesis. Chapter 2is of preliminary character. We recall the concept of Pareto minimal pointsand some notions related to set-valued maps and variational inequality pro-blem. We introduce some scalarizing functions, in particular the so-calledaugmented biggest/smallest monotone functions and augmented signed distance functions, and establish some properties we shall use later.Chapter 3 describes the traffic network models to be studied in this thesis.We define equilibrium for each model and determine a relationship betweenthem. We also give some counter examples for some existing results in therecent literature on this topic.In Chapter 4 we develop a new solution method for multi-criteria net-work equilibrium problems without capacity constraints. To this end we shallconstruct two optimization problems the solutions of which are exactly theset of equilibria of the model, and establish some important generic conti-nuity and differentiability properties of the objective functions. Then we givethe formula to calculate the gradient of the objective functions which enablesus to modify Frank-Wolfe's reduced gradient method to get descent directiontoward an optimal solution. We prove the convergence of the method whichgenerates a nice representative set of equilibria. Since the objective functionsof our optimization problems are not continuous, a method of smoothingthem is also considered in order to see how global optimization algorithmsmay help.We shall also introduce the concept of robust equilibrium, establishcriteria for robustness and a formula to compute the radius of robustness.In Chapter 5 we consider vector equilibrium in the multi-criteria single-product traffic network with capacity constraints.We propose an equivalent optimization problem and establish some im-portant generic continuity and differentiability properties of the objectivefunction. Then we give a formula which allows us to calculate the gradientof the objective function. After that we apply the approach of Chapter 4 toobtain an algorithm for generating equilibria of this network. We also givesome numerical examples to illustrate our approach.In the last chapter we consider strong vector equilibrium in the multi-criteria multi-product traffic network with capacity constraints.We establish conditions for existence of strong vector equilibrium.We alsoestablish relations between equilibrium and efficient points of the value set ofthe cost function and with equilibrium with respect to a family of functions.Moreover we exploit particular increasing functions discussed in Chapter 2 toconstruct variational inequality problems, solutions of which are equilibriumflows. The final part of this chapter is devoted to an algorithm for findingequilibrium flows of a multi-criteria network with capacity constraints. Somenumerical examples are given to illustrate our method and its applicability.A list of references and appendices containing the code Matlab of ouralgorithms follow.

In this thesis, we study a numerical tool named multi-mesh method within the framework of the adaptive finite element method. The aim of this method is to minimize the size of the linear system to get the optimal performance of simulations. Multi-mesh methods are typically used in multi-physics problems, where more than one component is involved in the system. During the discretization of the weak formulation of partial differential equations, a finite-dimensional space associated with an independently refined mesh is assigned to each component respectively. The usage of independently refined meshes leads less degrees of freedom from a global point of view. To our best knowledge, the first multi-mesh method was presented at the beginning of the 21st Century. Similar techniques were announced by different mathematics researchers afterwards. But, due to some common restrictions, this method is not widely used in the field of numerical simulations. On one hand, only the case of two-mesh is taken into scientists\' consideration. But more than two components are common in multi-physics problems. Each is, in principle, allowed to be defined on an independent mesh. Besides that, the multi-mesh methods presented so far omit the possibility that coefficient function spaces live on the different meshes from the trial and test function spaces. As a ubiquitous numerical tool, the multi-mesh method should comprise the above circumstances. On the other hand, users are accustomed to improving the performance by taking the advantage of parallel resources rather than running simulations with the multi-mesh approach on one single processor, so it would be a pity if such an efficient method was only available in sequential. The multi-mesh method is actually used within local assembling process, which should not be conflict with parallelization. In this thesis, we present a general multi-mesh method without the limitation of the number of meshes used in the system, and it can be applied to parallel environments as well. Chapter 1 introduces the background knowledge of the adaptive finite element method and the pioneering work, on which this thesis is based. Then, the main idea of the multi-mesh method is formally derived and the detailed implementation is discussed in Chapter 2 and 3. In Chapter 4, applications, e.g. the multi-phase flow problem and the dendritic growth, are shown to prove that our method is superior in contrast to the standard single-mesh finite element method in terms of performance, while accuracy is not reduced.

Ce projet fait l’objet d’une collaboration entre l’équipe EPI du laboratoire SATIE de l’ENS Cachan, le laboratoire LAPLACE et ALSTOM Transport dans le cadre du projet ANR « CONCIGI HT ». Les recherches dans le domaine du ferroviaire visent à réduire toujours plus la taille et le poids de la chaine de traction. Aujourd’hui, la volonté de poursuivre la réduction des temps de parcours est complétée par le souhait de diminuer la puissance consommée, ceci ayant des conséquences importantes sur la conception des chaines de traction. L’objectif de ce projet est de remplacer certaines parties de la chaine de puissance d’une locomotive par des systèmes permettant de diminuer de 30% leur encombrement, donc de diminuer jusqu’à 8,5 % l’énergie électrique consommée et augmenter le nombre de passagers transportés. Le travail présenté dans cette thèse porte sur le dimensionnement et l’étude de l’alimentation des allumeurs de semi-conducteurs SiC HT (pouvant atteindre un transitoire de 10 kV). La particularité de ces semi-conducteurs est leur application : ils constituent des convertisseurs multi-niveaux dont la fonction est de remplacer les transformateurs 25 kV/3 kV-50 Hz actuellement utilisés pour la traction ferroviaire. La principale contrainte porte donc sur la tenue diélectrique qui atteint un maximum de 60 kV liée aux transitoires présents sur la caténaire. Une alimentation à double isolation galvanique (DGIT-Double Galvanic Insulation Transformer) a donc été développée afin de répondre à la nécessité de double tenue diélectrique (10 kV et 60 kV). Dans un premier temps une structure permettant la double isolation galvanique a été développée avec l’objectif de réduire au maximum sa taille et son poids. Sa disposition spatiale et sa géométrique ont également été pris en considération (ainsi que de nombreux paramètre à la fois pertinents et contradictoires). Ceci afin d’obtenir le système optimal et une répartition des capacités de mode commun et différentiel respectant la forte tenue diélectrique. Dans un second temps une étude de l’alimentation adaptée au DGIT a été réalisée et testée. L’association du DGIT qui est un élément très inductif, de la charge qui est un driver à faible puissance et de l’aspect multi-niveau implique un fonctionnement atypique. Pour chacune de ces parties, une étude structurelle, fréquentielle et électrique a été réalisée afin d’obtenir un dispositif optimal du point de vue du volume, du poids et des pertes, tout en respectant la principale contrainte de la tenue à la THT (Très Haute Tension). Enfin, dans un troisième temps nous avons étudié la possibilité de remplacer les fibres optiques utilisées de nos jours pour la commande des allumeurs par un système basé sur la transmission par radio fréquence. / This project is the result of collaboration between the EPI team of the SATIE laboratory at ENS Cachan, the LAPLACE laboratory and ALSTOM Transport in the ANR "CONCIGI HT" project. Research in the railway traction area aims to reduce the increasing size and weight of the power train. Today, the desire to reduce travel time is complemented by the desire to reduce power consumption- it is an important information for the design of traction chains. The objective of this project is to replace parts of the locomotive power supply systems to reduce their size by 30%, thus to reduce the power consumption and increase by 8.5% the number of passengers. The work presented in this thesis focuses on the design and study of the drivers power supply for HV SiC semiconductor (up to a 10 kV transient voltage). The peculiarity of these semiconductors is their application: they are placed in multi-level converters in order to replace the 25 kV / 3 kV, 50 Hz transformers currently used for rail traction. The main constraint is the dielectric strength that reaches a maximum of 60 kV due to transient present on the catenary. So, the Double Galvanic Insulation Transformer (DGIT) has been developed to adapt to the need of holding dual dielectric (10 kV and 60 kV). In a first step, a structure for the dual galvanic insulation has been developed with the objective to minimize its size and weight. Its spatial and geometrical arrangements were also taken into account (as well as many parameters both relevant and contradictory), in order to obtain the optimal system and common and differential mode distributed capacities observing the high dielectric strength. In a second step a study of the DGIT adapted power supply has been completed and tested. The combination of DGIT inductive behavior, the driver low power and the multi-level aspect, involves an atypical work of this power supply converter. For each of these parts, a structural, frequencial and electrical study was performed in order to achieve the maximum optimal device in terms of volume, weight and loss, with respect to the main constraint of VHV (Very High Voltage). Finally, in a third step we studied the possibility of subtituing the optical fibers currently used for the control of lighters with a system based on radio-frequency transmission.

Cette thèse porte sur trois questions qui se posent en optimisation multi-objectif. Dansun premier temps, nous étudions l’existence de solutions efficaces via des techniquesde scalarisation. On étend le théorème de Benson du cas convexe à un cas général.De plus, nous examinons d’autres techniques de scalarisation. Dans un second temps,nous abordons la question de robustesse. Nous examinons les concepts proposés dansla littérature sur le sujet. On étend au cas d’optimisation multi-objectif non-linéairela définition de Georgiev et ses collaborateurs. Quelques conditions nécessaires etsuffisantes pour obtenir une solution robuste moyennant des hypothèses appropriéessont données. Les relations entre cette notion de robustesse et certaines définitionsmentionnées sont mises en évidence. Deux types de modifications des fonctions objectifsont traités et les relations entre les solutions faibles/propres/ robustes efficacessont établies. Le dernier chapitre est consacré à l’analyse de sensibilité et de stabilitéen optimisation multi-objectif paramétrée. On montre sous des conditions faibles quela multi-application de l’ensemble des solutions réalisables et des valeurs réalisablessont strictement semi-différentiables. On donne quelques conditions suffisantes pourla semi-différentiabilité de l’ensemble efficace et des valeurs efficaces. De plus, nousétudions la pseudo-Lipschitz continuité des multi-applications ci dessus citées. / In this thesis, three crucial questions arising in multi-objective optimization are investigated.First, the existence of properly efficient solutions via scalarization toolsis studied. A basic theorem credited to Benson is extended from the convex caseto the general case. Some further scalarization techniques are also discussed. Thesecond part of the thesis is devoted to robustness. Various notions from the literatureare briefly reviewed. Afterwards, a norm-based definition given by Georgiev, Lucand Pardalos is generalized to nonlinear multi-objective optimization. Necessary andsufficient conditions for robust solutions under appropriate assumptions are given.Relationships between new robustness notion and some known ones are highlighted.Two kinds of modifications in the objective functions are dealt with and relationshipsbetween the weak/proper/robust efficient solutions of the problems, before and afterthe perturbation, are established. Finally, we discuss the sensitivity analysis andstability in parametrized multi-objective optimization. Strict semi-differentiability ofset-valued mappings of feasible sets and feasible values is proved under appropriateassumptions. Furthermore, some sufficient conditions for semi-differentiability of efficientsets and efficient values are presented. Finally, pseudo-Lipschitz continuity ofaforementioned set-valued mappings is investigated

Modelos multi-estados têm sido utilizados para descrever o comportamento de unidades amostrais cuja principal resposta é o tempo necessário para a ocorrência de seqüências de eventos. Consideramos um modelo multi-estados markoviano, não homogêneo, que incorpora covariáveis cujos efeitos podem variar ao longo do tempo (efeitos dinâmicos), o que permite a generalização dos modelos usualmente empregados. Resultados assintóticos mostram que procedimentos de estimação baseados no método histograma crivo convergem para um processo gaussiano. A metodologia proposta mostra-se adequada na modelagem de dados reais para comparação de desenvolvimento de recém-nascidos pré-termo com os a termo. Estudos com dados gerados artificialmente confirmam os resultados teóricos obtidos. / Multi-state models have been used to describe the behavior of sample units where the principal response is the time needed for the occurrence of a sequence of events. We consider a non-homogeneous Markovian multi-state model that incorporates covariates with time-dependent coefficient (dynamic effects), generalizing models usually employed. The asymptotic results show that the estimators based on the method of histogram sieves converge to a Gaussian process. The proposed methodology revels adequated for modeling data related to the comparison of developement of preterm infants with term infants. The studies with artificially generated data confirm the asymptotic results.

Esta tese é dedicada ao estudo de modelos de otimização de carteiras de investimento multi-período. Daremos ênfase a um modelo com restrições intermediárias formulado como um problema de controle ótimo e resolvido utilizando técnicas de programação dinâmica. Serão tratados aspectos teóricos e práticos desta classe de problemas. Primeiramente faremos uma revisão das principais hipóteses dos modelos de otimização de carteiras e o caso uni-período. Analisaremos a seguir as generalizações para o caso multi-período, onde os modelos utilizam apenas restrições para o valor esperado e/ou para a variância da carteira no instante final do período analisado. Apresentaremos então o principal resultado proposto neste trabalho onde consideramos o problema de seleção ótima de ativos multi-período no qual podemos incorporar ao modelo restrições intermediárias para o valor esperado e variância da carteira durante o período de análise. A grande vantagem desta técnica é permitir o controle do valor esperado e/ou da variância da carteira ao longo de todo o horizonte de análise. Faremos uma comparação o entre as formulações apresentadas e realizaremos experimentos numéricos com o modelo proposta nesta tese. Os principais resultados originais desta tese encontram-se no Capítulo 5. No Capítulo 6 apresentamos as simulações numéricas realizadas com o modelo proposto. / The subject of this thesis is the study of multi-period portfolio optimization problems. We focus on a model with intermediate constraints formulated as an optimal control problem and solved by using dynamic programming techniques. Both theoretical and practical issues are addressed. Firstly we will analyze the main hypothesis of portfolio optimization models and the single period case. Then we will present the generalization for the multi-period case, where the models use only constraints for the expected value and variance at the final period. The main result proposed in this work considers the multi-period portfolio selection problem with intermediate constraints on the expected value and variance of the portfolio taken into account in the optimization problem. The main advantage of this technique is that it is possible to control the intermediate expected value or variance of the portfolio during the time horizon considered. Comparison between the presented formulations and numerical experiments of the proposed model will be exposed. The main original results of this thesis can be found in Chapter 5. In Chapter 6 we present numerical simulations with the proposed model.

Visando contribuir com o estado-da-arte de sistemas bioinspirados em formigas na robóotica, neste trabalho é abordado o problema do controle de um grupo de robôs para a solução coletiva das tarefas de exploração do ambiente e localização de objetos. Para isso, são utilizados algoritmos inspirados em colônias de formigas. O objetivo deste trabalho, portanto, é o desenvolvimento de um sistema de controle de navegação baseado em colônia de formigas para um time de robôs, de maneira que os robôs resolvam esses problemas utilizando estratégias de controle individuais e simples. Esse sistema tem como base a utilização de marcadores ou feromônios artificiais, que podem ser depositados pelos robôs para marcar determinadas posiçôes do ambiente / Aiming to advance the state-of-the-art of ant bioinspired systems in robotic applications, in this work we study the problem of controling a group of robots for solving colective tasks on environment exploration and object localization. To this end, we used algorithms inspired in ant colonies. Therefore, the objective of this work is to develop a navigation control system based on ant colony can solve the problems using simple control strategies. This system uses marks or artificial pheromones that can be released by the robots to mark specific positions in the environment

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / Devido ao rÃpido crescimento e os agressivos requisitos de vazÃo nas atuais redes sem fio, como os sistemas celulares de 4 a GeraÃÃo, a interferÃncia se tornou um problema que nÃo pode mais ser negligenciado. Neste contexto, o Alinhamento de InterferÃncia (IA) tem surgido como uma tÃcnica promissora que possibilita transmissÃes livres de interferÃncia com elevada eficiÃncia espectral. No entanto, trabalhos recentes tÃm focado principalmente nos ganhos teÃricos que esta tÃcnica pode prover, enquanto esta dissertaÃÃo visa dar um passo na direÃÃo de esclarecer alguns dos problemas prÃticos de implementaÃÃo da tÃcnica em redes celulares, bem como comparÃ-la com outras tÃcnicas bem estabelecidas. Uma rede composta por trÃs cÃlulas foi escolhida como cenÃrio inicial de avaliaÃÃo, para o qual diversos fatores realistas foram considerados de modo a realizar diferentes anÃlises. A primeira anÃlise foi baseada em imperfeiÃÃes de canal, cujos resultados mostraram que o IA à mais robusto aos erros de estimaÃÃo de canal que o BD (do inglÃs, Block Diagonalization), enquanto as duas abordagens sÃo igualmente afetadas pela correlaÃÃo entre as antenas. O impacto de uma interferÃncia externa nÃo-coordenada, que foi modelada por diferentes matrizes de covariÃncia de modo a emular vÃrios cenÃrios, tambÃm foi avaliado. Os resultados mostraram que as modificaÃÃes feitas nos algoritmos de IA podem melhorar bastante seus desempenho, com uma vantagem para o algoritmo que suprime um Ãnico fluxo de dados, quando sÃo comparadas as taxas de erro de bit alcanÃadas por cada um. Para combinar os fatores das anÃlises anteriores, as variaÃÃes temporais de canal foram consideradas. Neste conjunto de simulaÃÃes, alÃm da presenÃa da interferÃncia externa, os prÃ-codificadores sÃo calculados atravÃs de medidas atrasadas de canal, levando a resultados que corroboraram com as anÃlises anteriores. Um fato recorrente percebido em todas as anÃlises anteriores à o dilema entre aplicar os algoritmos baseados em BD, para que se consiga alcanÃar maiores capacidades, ou enviar a informaÃÃo atravÃs de um enlace mais confiÃvel utilizando o IA. Uma maneira de esclarecer este dilema à efetivamente realizar simulaÃÃes a nÃvel sistÃmico, para isto foi aplicado um simulador sistÃmico composto por um grande nÃmero de setores. Como resultado, todas as anÃlises realizadas neste simulador mostraram que a tÃcnica de IA atinge desempenhos intermediÃrios entre a nÃo cooperaÃÃo e os algoritmos baseados na prÃ-codificaÃÃo conjunta. Uma das principais contribuiÃÃes deste trabalho foi mostrar alguns cenÃrios em que a tÃcnica do IA pode ser aplicada. Por exemplo, quando as estimaÃÃes dos canais nÃo sÃo tÃo confiÃveis à melhor aplicar o IA do que os esquemas baseados no processamento conjunto. TambÃm mostrou-se que as modificaÃÃes nos algoritmos de IA, que levam em consideraÃÃo a interferÃncia externa, podem melhorar consideravelmente o desempenho dos algoritmos. Finalmente, o IA se mostrou uma tÃcnica adequada para ser aplicada em cenÃrios em que a interferÃncia à alta e nÃo à possÃvel ter um alto grau de cooperaÃÃo entre os setores vizinhos. / Due to the rapid growth and the aggressive throughput requirements of current wireless networks, such as the 4th Generation (4G) cellular systems, the interference has become an issue that cannot be neglected anymore. In this context, the Interference Alignment (IA) arises as a promising technique that enables transmissions free of interference with high-spectral efficiency. However, while recent works have focused mainly on the theoretical gains that the technique could provide, this dissertation aims to go a step further and clarify some of the practical issues on the implementation of this technique in a cellular network, as well as compare it to other well-established techniques. As an initial evaluation scenario, a 3-cell network was considered, for which several realistic factors were taken into account in order to perform different analyses. The first analysis was based on channel imperfections, for which the results showed that IA is more robust than Block Diagonalization (BD) regarding the Channel State Information (CSI) errors, but both are similarly affected by the correlation among transmit antennas. The impact of uncoordinated interference was also evaluated, by modeling this interference with different covariance matrices in order to mimic several scenarios. The results showed that modifications on the IA algorithms can boost their performance, with an advantage to the approach that suppresses one stream, when the Bit Error Rate (BER) is compared. To combine both factors, the temporal channel variations were taken into account. At these set of simulations, besides the presence of an external interference, the precoders were calculated using a delayed CSI, leading to results that corroborate with the previous analyses. A recurring fact on the herein considered analyses was the dilemma of weather to apply the Joint Processing (JP)-based algorithms in order to achieve higher sum capacities or to send the information through a more reliable link by using IA. A reasonable step towards solving this dilemma is to actually perform the packet transmissions, which was accomplished by employing a system-level simulator composed by a large number of Transmission Points (TPs). As a result, all analyses conducted with this simulator showed that the IA technique can provide an intermediate performance between the non-cooperation and the full cooperation scheme. Concluding, one of the main contributions of this work has been to show some scenarios/cases where the IA technique can be applied. For instance, when the CSI is not reliable it can be better to use IA than a JP-based scheme. Also, the modifications on the algorithms to take into account the external interference can boost their performance. Finally, the IA technique finds itself in-between the conventional transmissions and Coordinated Multi-Point (CoMP). IA achieves an intermediate performance, while requiring a certain degree of cooperation among the neighboring sectors, but demanding less infrastructure than the JP-based schemes.

nÃo hà / Com as crescentes demandas pela provisÃo de serviÃos de dados em redes celulares, seus operadores tem arcado com os problemas de acomodar as crescentes capacidades enquanto mantÃm os nÃveis de qualidade de serviÃo. As redes de Multi-Acesso (MA) constituem uma maneira eficiente de oferecer estas capacidades, jà que oferecem maneiras de agregar os recursos de rÃdio das Tecnologias de Acesso de Radio (RATs)1 empregadas. Apesar do conceito de Multi-Acesso ser interessante, os ganhos providos por sua adoÃÃo nÃo sÃo claros. Tais ganhos podem vir de diferentes fontes, tais como, a inclusÃo de uma nova RAT e do emprego de procedimentos de SeleÃÃo de Acesso (AS) adequados. A primeira fonte diz respeito aos ganhos disponibilizados pelo acrÃscimo de recursos de rÃdio, enquanto a segunda diz respeito à melhor distribuiÃÃo dos usuÃrios as RATs. Este trabalho visa tornar claro o ganho envolvido na adiÃÃo de uma Rede Local Sem Fio (WLAN) 802.11a a uma UMTS Terrestrial Radio Access Network (UTRAN), tanto para uma distribuiÃÃo de usuÃrios homogÃnea quanto para uma situaÃÃo de hotspot, empregando o conceito de redes de MA como tecnologia habilitadora. TambÃm entra em detalhes sobre as caracterÃsticas de alguns procedimentos de AS2 baseados em cobertura, taxa de transmissÃo do usuÃrio, vazÃo estimada da RAT e mista (taxa de transmissÃo e carga), tentando apontar quais fornecem os melhores desempenhos. A avaliaÃÃo à baseada em simulaÃÃes dinÃmicas e sistÃmicas que contam com modelagem detalhada das tecnologias e conceitos envolvidos. Os resultados mostram que, considerando a modelagem e prÃ-suposiÃÃes, ganhos significativos podem ser conseguidos usando algoritmos de AS inteligentes no cenÃrio do MA envolvendo as redes UTRAN e WLAN. / With the increasing demands for data services over the cellular networks, their operators are faced with the issue of providing the requested capacities while maintaining the service quality levels. The Multi-Access (MA) networks can be a cost effective way of raising those capacities, as it provides means to aggregate radio resources from the adopted Radio Access technologies (RATs) involved. Although the MA concept seems very promising, the gains concerning its adoption are not quite clear. These gains can come from different sources, such as from the inclusion of a second RAT itself and from the adoption of a suitable Access Selection (AS) procedure. The former concerns the gains available with the increase in the amount of radio resources while the latter concerns the ones available with a more suitable distribution of users among the RATs. This work aims to clarify the gains concerning the usage of the WLAN IEEE 802.11a in addition to the UTRAN, both in the homogeneous user distribution and hotspot cases, while employing the MA framework as the enabling technology. It also looks into the characteristics of some AS procedures based on coverage, user bit rate, estimated RAT throughput and mixed (user bit rate and load), aiming at clarifying which ones provide the best performance. The evaluation is based on a dynamic system-level multi-access simulator, comprised of a good and detailed level of the modeling concerning the involved technologies and concepts. Results show that considering the assumptions and modeling, significant capacity gains can be achieved when using intelligent AS algorithms in the UTRAN-WLAN MA networks.

O desenvolvimento de arquiteturas de controle para múltiplos robôs em ambientes dinâmicos tem sido tema de pesquisas na área de robótica. A complexidade deste tema varia de acordo com as necessidades exigidas da equipe de robôs. Em geral, espera-se que os robôs colaborem uns com os outros na execução de uma tarefa. Além disso, cada robô deve ser capaz de planejar trajetórias e replanejá-las em caso de situações inesperadas. No presente trabalho, propomos uma Arquitetura de Controle Inteligente para múltiplos robôs denominada ACIn. Para esta finalidade, foram investigadas algumas técnicas utilizadas para o controle inteligente de robôs, tais como, Redes Neurais Artificiais, Campos Potenciais e Campos Potenciais baseados em Problema do Valor de Contorno (PVC). Tais técnicas, normalmente utilizadas para um único robô, foram adaptadas para tornar possível o controle de múltiplos robôs sob arquitetura ACIn. Uma outra contribuição deste trabalho refere-se ao aperfeiçoamento da técnica de Campos Potenciais baseada PVC denominada Campos Potenciais Localmente Orientados (CPLO). Este aperfeiçoamento foi proposto para suprir a deficiência das técnicas baseadas em PVC quando estas são aplicadas em ambientes com múltiplos robôs. Além disso, um Sistema Baseado em Regras (SBR) também foi proposto como parte integrante da arquitetura ACIn. O objetivo do SBR é caracterizar a funcionalidade de cada robô para uma determinada tarefa. Isto se faz necessário para que o comportamento dos integrantes da equipe de robôs não seja competitivo e sim colaborativo. Por fim, através dos experimentos utilizando o simulador oficial de futebol de robôs da FIRA, observou-se que a arquitetura de controle inteligente (ACIn) implementada com a técnica de planejamento CPLO e SBR propostos, mostrou-se robusta e eficiente no controle de múltiplos robôs / In this work, an intelligent control architecture for multi-robots denominated ACIn was proposed. With this objective, some techniques considered intelligent were studied for the planning of trajectories, such as Artificial Neural Networks, Potential Fields and Potential Fields based on the boundary value problem (BVP). Such techniques, normally used for a single robot, were adapted to function with multi-robots inside the ACIn architecture. Another contribution of this work refers to the improvement of the Potential Fields based on the boundary value problem (BVP) technique. This improvement was proposed to supply the drawback of the BVP based techniques when they are applied to multi robots environments. Besides, a Rule Based System (RBS) was also proposed as part of the ACIn architecture. The objective of the RBS is to characterize the functionality of each robot for a determined task. This is necessary for the behavior of the equip members not to be competitive, but collaborative. Finally, it was observed through the experiments with the robot soccer simulated environment, that our intelligent control architecture (ACIn) proposal, integrating planning and RBS for the control of multi-robots was satisfactory