We present a system architecture and a set of control techniques that allow heterogeneous software design tools to collaborate intelligently and automatically. One of their distinguishing features is the ability to perform concurrent processing. Systems based on this architecture are able to effectively solve large electromagnetic analysis problems, particularly those that involve loose coupling between several areas of physics. The architecture can accept any existing software analysis tool, without requiring any modification or customization of the tool. This characteristic is produced in part by our use of a neutral virtual representation for storing problem data, including geometry and material definitions. We construct a system based on this architecture, using several circuit and finite-element analysis tools, and use it to perform electromagnetic analyses of several different devices. Our results show that our architecture and techniques do allow practical problems to be solved effectively by heterogeneous tools. / On présente une architecture de système et un ensemble de techniquesde contrôle qui permettent aux logiciels d'analyse hétérogènes de collaborerde façon intelligente et automatique. Un de ses traits caractéristiques est sacapacité d'effectuer simultanément plusieurs traitements. Les systèmes baséssur cette architecture sont capables de résoudre de manière efficace des grandsproblèmes dans le domaine de l'analyse électromagnétique, particulièrementceux où existe un accouplement dégagé entre plusieurs domaines de physique.L'architecture peut accepter n'importe quel logiciel d'analyse existant; ellen'exige pas que les logiciels soyent modifiés ou fabriqués sur mesure. Cettecaractéristique est produite en partie par notre utilisation d'une représentationneutre virtuelle pour représenter les données du problème, y inclus sa géométrieet les proprietés de ses matériels. On construit un système basé sur cettearchitecture, comprenant plusieurs logiciels de simulation, et on l'emploie pourexécuter des analyses électromagnétiques de plusieurs appareils différents. Nosrésultats montrent que notre architecture et nos techniques permettent desproblèmes pratiques d'être résolus efficacement par les outils hétérogènes.

In networked robot systems (NRS), robots and robotic devices are distributed in the environment; typically tasks are performed by cooperation and coordination of such multiple networked components. NRS offer advantages over monolithic systems in terms of modularity, flexibility and cost effectiveness, and they are thus becoming a mainstream approach to the inclusion of robotic solutions in everyday environments. The components of a NRS are usually robots and sensors equipped with rich computational and communication facilities. In this thesis, we argue that the capabilities of a NRS would greatly increase if it could also accommodate among its nodes simpler entities, like small ubiquitous sensing and actuation devices, home appliances, or augmented everyday objects. For instance, a domestic robot needs to manipulate food items and interact with appliances. Such a robot would benefit from the ability to exchange information with those items and appliances in a direct way, in the same way as with other networked robots and sensors. Combining such highly heterogeneous devices inside one NRS is challenging, and one of the major challenges is to provide a common communication and collaboration infrastructure. In the field of NRS, this infrastructure is commonly provided by a shared middleware. Unfortunately, current middlewares lack the generality needed to allow heterogeneous entities such as robots, simple ubiquitous devices and everyday objects to coexist in the same system. In this thesis we show how an existing middleware for NRS can be extended to include three new types of “citizens” in the system, on peer with the other robots. First, we include computationally simple embedded devices, like ubiquitous sensors and actuators, by creating a fully compatible tiny version of the existing robotic middleware. Second, we include augmented everyday objects or home appliances which are unable to run the middleware on board, by proposing a generic design pattern based on the notion of object proxy. Finally,we go one step further and include humans as nodes in the NRS by defining the notion of human proxy. While there exist a few other NRS which are able to include both robots and simple embedded devices in the same system, the use of proxies to include everyday objects and humans in a generic way is a unique feature of this work. In order to verify and validate the above concepts, we have implemented them in the Peis-Ecology NRS model. We report a number of experiments based on this implementation, which provide both quantitative and qualitative evaluations of its performance, reliability, and interoperability.

The application of static shape control was investigated in this thesis particularly for a composite plate configuration using piezoelectric actuators. A new electro-mechanically coupled mathematical model was developed for the analysis and is based on a third order displacement field coupled with a layerwise electric potential concept. This formulation, TODL, is then implemented into a finite element program. The mathematical model represents an improvement over existing formulations used to model intelligent structures using piezoelectric materials as actuators and sensors. The reason is TODL does not only account for the electro-mechanical coupling within the adaptive material, it also accounts for the full structural coupling in the entire structure due to the piezoelectric material being attached to the host structure. The other significant improvement of TODL is that it is applicable to structures which are relatively thick whereas existing models are based on thin beam / plate theories. Consequently, transverse shearing effects are automatically accounted for in TODL and unlike first order shear deformation theories, shear correction factors are not required. The second major section of this thesis uses the TODL formulation in static shape control. Shape control is defined here as the determination of shape control parameters, including actuation voltage and actuator orientation configuration, such that the structure that is activated using these parameters will conform as close as possible to the desired shape. Several shape control strategies and consequently algorithms were developed here. Initial investigations in shape control has revealed many interesting issues which have been used in later investigations to improve shape controllability and also led to the development of improved algorithms. For instance, the use of discrete actuator patches has led to greater shape controllability and the use of slopes and curvatures as additional control criteria have resulted in significant reduction in internal stresses. The significance of optimizing actuator orientation and its relation to piezoelectric anisotropy in improving shape controllability has also been presented. Thus the major facets of shape control has been brought together and the algorithms developed here represent a comprehensive strategy to perform static shape control.

Thesis (M.S.)--Air Force Institute of Technology, 2008. / Title from title page of PDF document (viewed on: Dec 10, 2009).

Το παρόν τεύχος αποτελεί διπλωματική εργασία που αναπτύχθηκε στο Εργαστήριο Γενικής Ηλεκτροτεχνίας του Τομέα Συστημάτων και Αυτόματου Ελέγχου. Συνίσταται στην κατασκευή και λειτουργία μιας πειραματικής διάταξης αυτοματισμού στην οποία εφαρμόζονται τεχνικές ελέγχου για την πραγματοποίηση ελεγχόμενης αεροδυναμικής αιώρησης ελαφρών σωμάτων. Περιλαμβάνει εξοπλισμό όπως αναλογικές βαλβίδες πεπιεσμένου αέρα, αισθητήρες και έναν Προγραμματιζόμενo Λογικό Ελεγκτή (PLC) που ελέγχει τη λειτουργία της διάταξης. Η πειραματική διάταξη αυτοματισμού που κατασκευάστηκε είναι ένα σύστημα αιώρησης ελαφρών σωμάτων με χρήση πεπιεσμένου αέρα. Αισθητήρες θέσης προσφέρουν την δυνατότητα γνώσης κάθε χρονική στιγμή της θέσης του σώματος. Μέσω του Προγραμματιζόμενου Λογικού Ελεγκτή επιτυγχάνουμε αιώρηση του σώματος σε αυστηρά ελεγχόμενο ύψος κάνοντας χρήση διαφόρων μεθόδων ελέγχου (On-Off, PID). Τα αντικείμενα με τα οποία πειραματιστήκαμε ήταν κυρίως σφαίρες μικρής διαμέτρου(4-5cm) και βάρους μερικών γραμμαρίων. Κατά την διάρκεια εκπόνησης της διπλωματικής, έγινε προσπάθεια αιώρησης και ελέγχου επίπεδων αντικειμένων. Με το διατιθέμενο εξοπλισμό (περιορισμοί πίεσης αέρα, είδους ακροφυσίου κλπ.) κατέστη δυνατή η αιώρηση επίπεδων αντικειμένων μόνο για μικρή χρονική διάρκεια. Δοκιμάσθηκαν διάφοροι τύποι και χωρικοί συνδυασμοί ακροφυσίων για τη δημιουργία επίπεδου στρώματος “πεπιεσμένου” αέρα. Στην πράξη διαπιστώθηκε ταλαντωτική συμπεριφορά των επίπεδων αντικειμένων και ισχυρή αστάθεια του όλου συστήματος. Στο κεφάλαιο 1 παρουσιάζονται οι επικρατέστερες τεχνικές αιώρησης. Στο κεφάλαιο 2 προσεγγίζουμε θεωρητικά την αεροδυναμική αιώρηση παρουσιάζοντας τις σημαντικότερες αρχές και νόμους που περιγράφουν το φαινόμενο αυτό. Στο κεφάλαιο 3 υπάρχει μία εκτενής ανάλυση των ηλεκτροπνευματικών συστημάτων και στην συνέχεια των Προγραμματιζόμενων Λογικών Ελεγκτών (PLC) καθώς και το λογισμικό επικοινωνίας με PLC, το Step 7. Στο κεφάλαιο 4 αναλύεται ο εξοπλισμός της πειραματικής διάταξης, περιγράφονται τα πειράματα που διεξάγαμε και παρουσιάζεται και αναλύεται το πρόγραμμα που υλοποιήθηκε σε γλώσσα STL. Στο κεφάλαιο 5 συνοπτικά παρουσιάζεται το πρόγραμμα PLC Analyzer με το οποίο καταγράψαμε όλες τις πειραματικές μετρήσεις της διάταξης αιώρησης και τέλος παραθέτονται τα διαγράμματα από κάθε μία μέτρηση. / --

O principal objetivo do controle de vibrações é a sua redução ou minimização, através da modificação automática da resposta estrutural. Em muitas situações isto é necessário para promover a estabilidade estrutural, e para alcançar o alto desempenho mecânico necessário em diversas áreas técnicas, tais como a engenharia aeroespacial, civil e mecânica, bem como a biotecnologia, inclusive em escala micro e nano mecânica. Uma alternativa é o uso de estruturas inteligentes, que são o resultado da combinação de sensores e atuadores integrados em uma estrutura mecânica, e um método de controle adequado. O principal objetivo deste trabalho é o desenvolvimento de rotinas computacionais para a simulação, via método dos elementos finitos, do controle ativo de estruturas inteligentes de cascas, placas e vigas delgadas de material compósito laminado com camadas de material piezoelétrico como sensores e/ou atuadores. Caracterizam esta pesquisa a utilização do elemento GPL-T9 de três nós e seis graus de liberdade mecânicos por nó, mais um grau de liberdade elétrico por camada piezoelétrica, assim como a avaliação de dois métodos de controle, o Proporcional-Integral-Derivativo (PID) e o Regulador Quadrático Linear ou Linear Quadratic Regulator (LQR), incluindo o LQR Modal, e a otimização da localização de pastilhas piezoelétricas através de um Algoritmo Genético (AG). Várias aplicações são apresentadas e os resultados obtidos são comparados aos disponíveis na literatura especializada. / The main objective of vibration control is its reduction or even its minimization by the automatic modification of the structural response. Sometimes this is necessary to increase structural stability and to attain a high mechanical behavior in several areas such as aerospace, civil and mechanical engineering, biotechnology, including macro, micro and nanomechanical scales. An alternative is to use a smart structure, which results of the combinations of integrated sensors and actuators in a mechanical structure and a suitable control method. Development of a computational code to simulate, using finite elements, the active control in smart structures such as slender shells, plates and beams of composite materials with embedded piezoelectric layers acting as actuators and sensors is the main objective of this work. This research is characterized by the use of the GPL-T9 element with three nodes and six mechanical degrees of freedom and one electrical degree of freedom per piezoelectric layer, by the evaluation of two control methods, the Proportional Integral Derivative (PID) and the Linear Quadratic Regulator (LQR), including the Modal LQR, and, finally by the optimization of piezoelectric patches placement using a Genetic Algorithm (GA). Several examples are presented and compared with those obtained by other authors.

La sûreté de fonctionnement est devenue indispensable pour tous les systèmes critiques où des vies humaines sont en jeu (comme l’aéronautique, le ferroviaire, etc.). Ceci a conduit à la conception et au développement des architectures tolérantes aux fautes, l’objectif de ces architectures étant de maintenir un service correct délivré par le système malgré la présence de fautes, et en particulier de garantir la sécurité-innocuité et la fiabilité du système. La tolérance aux fautes sur les drones aériens multirotors a récemment reçu une attention importante de la part de la communauté scientifique. En particulier, plusieurs travaux ont été développés sur la tolérance aux fautes des quadrirotors suite à des fautes partielles sur les actionneurs, et récemment des recherches ont abordé le problème de panne totale de l’un des actionneurs. D’après ces études, il a été montré qu’une défaillance totale d’un actionneur dans un quadrirotor rend le système non complètement contrôlable. Une solution proposée est de considérer des multirotors avec des actionneurs redondants (hexarotors ou octorotors). La redondance inhérente disponible dans ces véhicules est exploitée, en cas de défaillance sur les actionneurs, pour redistribuer les efforts de commande sur les moteurs sains de façon à garantir la stabilité et la contrôlabilité complète du système. Dans ce travail de thèse, des approches pour la conception de systèmes de commande tolérants aux fautes des drones multirotors sont étudiées et appliquées au contrôle des octorotors. Toutefois, les algorithmes sont conçus de manière à être applicables sur les autres types de multirotors avec des modifications mineures. D’abord, une analyse de contrôlabilité de l’octorotor après l’occurrence de défaillances sur les actionneurs est présentée. Ensuite, un module de détection et d’isolation de pannes moteurs basé sur un observateur non-linéaire et les mesures de la centrale inertielle est proposé. Les mesures des vitesses et des courants de moteurs fournis par les contrôleurs de vitesse sont également utilisées dans un autre algorithme de détection pour détecter les défaillances des actionneurs et distinguer les pannes moteurs des pertes des hélices. Un module de rétablissement basé sur la reconfiguration du multiplexage est proposé pour redistribuer les efforts de commande d’une manière optimale sur les actionneurs sains après l’occurrence de défaillances dans le système. Une architecture complète, comprenant la détection et l’isolation des défauts suivie par le rétablissement du système est validée expérimentalement sur un octorotor coaxial puis elle est comparée à d’autres architectures basées sur l’allocation de commande et la tolérance aux fautes passive par mode glissant. / With growing demands for safety and reliability, and an increasing awareness about the risks associated with system malfunction, dependability has become an essential concern in modern technological systems, particularly safety-critical systems such as aircrafts or railway systems. This has led to the design and development of fault tolerant control systems (FTC). The main objective of a FTC architecture is to maintain the desirable performance of the system in the event of faults and to prevent local faults from causing failures. The last years witnessed many developments in the area of fault detection and diagnosis and fault tolerant control for Unmanned Aerial rotary-wing Vehicles. In particular, there has been extensive work on stability improvements for quadrotors in case of partial failures, and recently, some works addressed the problem of a quadrotor complete propeller failure. However, these studies demonstrated that a complete loss of a quadrotor motor results in a vehicle that is not fully controllable. An alternative is then to consider multirotors with redundant actuators (octorotors or hexarotors). Inherent redundancy available in these vehicles can be exploited, in the event of an actuator failure, to redistribute the control effort among the remaining working actuators such that stability and complete controllability are retained. In this thesis, fault-tolerant control approaches for rotary-wing UAVs are investigated. The work focuses on developing algorithms for a coaxial octorotor UAV. However, these algorithms are designed to be applicable to any redundant multirotor under minor modifications. A nonlinear model-based fault detection and isolation system for motors failures is constructed based on a nonlinear observer and on the outputs of the inertial measurement unit. Motors speeds and currents given by the electronic speed controllers are also used in another fault detection and isolation module to detect actuators failures and distinguish between motors failures and propellers damage. An offline rule-based reconfigurable control mixing is designed in order to redistribute the control effort on the healthy actuators in case of one or more motors failures. A complete architecture including fault detection and isolation followed by system recovery is tested experimentally on a coaxial octorotor and compared to other architectures based on pseudo-inverse control allocation and a robust controller using second order sliding mode.

L’objectif de cet projet est la conception, réalisation et caractérisation d’un actionneur / capteur piézoélectrique piézorésonant destiné à la mesure du vieillissement de la peau humaine. L’étude présentée est le fruit d’une collaboration entre le groupe de recherche de l'Electrodynamique du INP-ENSEEIHT (Toulouse), LAPLACE Laboratoire de Recherche et l'École Polytechnique de Gdask, Département Génie Electrique et Automatique. Un concept d’actionneur / capteur pour la caractérisation des propriétés mécaniques des tissus mous a été présenté. Un actionneur piézoélectrique résonant, appelé "unimorphe" a été choisi parmi les différentes structures piézoélectriques fondées sur le cahier des charges. L'innovation du projet réside dans l'intégration de la méthode d'indentation dynamique en utilisant un unimorphe comme dispositif d'indentation. Ceci permet l'utilisation d'un certain nombre de propriétés électromécaniques favorables des transducteurs piézo-électriques. Ce mémoire est divisé en 7 chapitres. Le chapitre 1 présente la thèse et ses objectifs. Le chapitre 2 présente le phénomène piézoélectrique et les applications piézoélectriques dans les domaines de la médecine et de la bio ingénierie. Le chapitre 3 décrit le cahier des charges pour le transducteur développé. Le choix du transducteur unimorphe est ainsi justifié. Le chapitre 4 présente une description analytique du transducteur unimorphe, y compris les calculs de déformations statiques, la description du circuit équivalent de Mason, et la description des conditions de contact entre la sonde d'indentation et les matériaux testés. Le chapitre 5 contient l'analyse numérique du transducteur unimorphe en utilisant le modèle virtuel MEF. Les résultats de simulations statiques et modales sont décrits par deux géométries considérées du transducteur. Le chapitre 6 décrit le processus de vérification expérimentale des modèles analytiques et numériques développés pour le transducteur unimorphe. Enfin, le dernier chapitre comprend des conclusions générales concernant les résultats de recherche obtenus, ainsi que les travaux futurs possibles. Afin de vérifier la thèse d'un cycle complet de recherche a été effectuée, qui a couvert: étude analytique, l'analyse numérique (simulations MEF), réalisation de prototype, et la vérification expérimentale des actionneurs / capteurs piézoélectriques considérés. / The main goal of the dissertation was following: preparation of a new concept, implementation and analysis of the piezoelectric resonant sensor/actuator for measuring the aging process of human skin. The research work has been carried out in the framework of cooperation between the INP-ENSEEIHT-LAPLACE, Toulouse, France, and at the Gdansk University of Technology, Faculty of Electrical and Control Engineering, Research Group of Power Electronics and Electrical Machines, Gdask, Poland. A concept of transducer for the characterization of mechanical properties of soft tissues was presented. The piezoelectric resonant, bending transducer, referred to as “unimorph transducer” was chosen from different topologies of piezoelectric benders based on the fulfillment of the stated requirements. The innovation of the project lies in the integration of the dynamic indentation method by using a unimorph as an indentation device. This allows the use of a number of attractive electromechanical properties of piezoelectric transducers. The thesis is divided into seven chapters. Chapter 1 states the thesis and goals of the dissertation. Chapter 2 presents piezoelectric phenomenon and piezoelectric applications in the fields of medicine and bioengineering. Chapter 3 describes the requirements for the developed transducer. The choice of unimorph transducer is justified. Chapter 4 presents an analytical description of the unimorph transducer, including the calculations of static deformations, equivalent circuit description, and description of the contact conditions between the transducer and the tested materials. Chapter 5 contains the numerical analysis of the unimorph transducer using FEM virtual model. Results of static and modal simulations are described for two considered geometries of the transducer. Chapter 6 describes the experimental verification process of analytic and numerical models developed for unimorph transducer. The final chapter includes general conclusions concerning obtained research results and achievements, as well as possible future works. In order to verify the proposition of the thesis a full research cycle was carried out, that covered: analytical study, numerical analysis (FEM simulations), prototype realization, and experimental verification of the considered (developed) piezoelectric sensor/actuator structures.

Com o aperfeiçoamento de materiais compósitos de alto desempenho, surge a possibilidade do desenvolvimento de estruturas inteligentes, onde atuadores e sensores piezelétricos estão integrados na estrutura com sistemas de controle adequados para a atenuação de vibrações. Projetos multidisciplinares se tornam cada vez mais complexos e sofisticados, envolvendo diversas fontes de incertezas que devem ser analisadas e quantificadas. O escopo principal desse trabalho visa o estudo da propagação de incertezas em estruturas de materiais compósitos laminados com atuadores e sensores piezelétricos, onde entradas e parâmetros do projeto podem ser fontes aleatórias e/ou nebulosas. Para atingir esse objetivo é adotada a metodologia fuzzy, com a aplicação de otimização de cortes-α. Essa técnica é utilizada na presença de informações vagas ou imprecisas acerca da aleatoriedade presente. Nesse estudo projetam-se, através do método dos elementos finitos, estruturas em forma de placa e casca de material compósito laminado com atuadores e sensores piezelétricos acoplados, controlados pelos reguladores Linear Quadratic Regulator (LQR) e Linear Quadratic Gaussian (LQG). Inicialmente são realizados estudos de otimização para encontrar a melhor localização dos componentes piezelétricos pelos Gramianos de controlabilidade e observabilidade, assim como os fatores de ponderação das leis de controle. O desenvolvimento é realizado no espaço modal reduzido visando um melhor desempenho computacional. As métricas escolhidas para avaliação do controle de vibração e análise das saídas incertas do sistema são baseadas nas energias cinética, potencial e elétrica. Também apresentam-se estudos de envelopes relacionados ao deslocamentos e às frequências naturais da estrutura devido às incertezas. Os resultados mostraram que as otimizações por corte-α para tratar números fuzzy nesse tipo de problema são robustas e eficientes, encontrando-se valores extremos das saídas desejadas. Além de ser um método não intrusivo, também pode ser utilizado em problemas com um número elevado de parâmetros incertos como entrada. / The possibility of developments of smart structures arises with high performance composite materials improvements, where piezoelectric actuators and sensors are embedded into the structures, following a suitable control laws for vibration attenuation. Multidisciplinary projects are becoming highly complex and sophisticated, involving several sources of uncertainty that should be analyzed and quantified. The main objective for this work is to study the uncertainty propagation in composite laminate structures with embedded piezoelectric actuators and sensors, considering random and/or fuzzy sources for the inputs and design parameters. To accomplish this objective, it is adopted the fuzzy α-cut optimizations methodology. This technique is used when the available information related to the actual randomness is vague or imprecise. In this study, laminated composite shells and plates structures are designed and analyzed by the finite element method, where embedded piezoelectric actuators and sensors controlled by Linear Quadratic Regulator (LQR) and Linear Quadratic Gaussian (LQG) are present. Initially, optimization analyses are executed to find the best arrangement for the piezoelectric material using controllability and observability Gramians metrics, as well as the best controller parameters. This study is developed in the reduced modal space looking for computational costs savings. The chosen rating metrics for the vibration control and uncertainty analysis are based on kinetic, potential and electrical energies. Structural displacements and natural frequency envelopes due uncertainty are also studied and presented. The results have shown that the fuzzy α-cut optimizations methodology is robust and efficient to find extreme values for the sought outputs. In addition to being a non-intrusive method, it is also able to deal with a large number of uncertain input parameters.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Wireless sensor and actuator networks are expected to grow quickly in size due to the decreasing cost of nodes. For such large scale networks, the geographic routing algorithm is a suitable approach. It finds a route for the packet based on the positions of the nodes. However, it often finds longer routes than the shortest possible ones, increasing the probability of packet loss. Moreover, the low cost wireless devices usually operate in the unlicensed bands, which is a crowded portion of the spectrum where several networks may be operating. Furthermore, the expected increase in the network density will also increase the competition for the wireless channel. Thus, intra and inter-network interferences play a major role in the reliability of the communication. In order to increase it, communication over multiple channels can be used. It exploits the capability of the radios that equip the nodes to be tuned over different frequencies. However, the number of available channels is limited in practice. Therefore, an allocation protocol must employ the channels eficiently. However, it was shown that to find an optimal allocation is NP-hard. Thus, one can use game theory to design a protocol that sub-optimally assigns channels in a distributed way. Game theory allows to model the problem as a game, where nodes are players and channels are their actions. According to the actions played, every node receives a payoff. The task of the protocol designer is to assign to every player a payoff function and a learning algorithm. This algorithm allows a player to play accordingly its observations of the game in order to maximize its payoff. If the designed game is potential, the algorithm can be used such that repeated playing converge to a steady state, known as Nash Equilibrium (NE). In a NE, no player can be better of by switching its current action and its received payoff can be bounded by a constant factor of the optimal payoff. By following such approach, a protocol called GBCA has been proposed, which allocates channels by exploiting topology and routing information. However, GBCA was designed for networks where routes are known in advance and static during a relatively long time. Geographic routing, by the other hand, finds the route for every packet on-demand. Therefore, in order to apply GBCA in networks that use geographic routing, this thesis proposes suitable modifications to the payoff functions. Using this modified protocol, which is called GBCA-G, it is shown, by simulations, the improvement of the network perfomance according to the metrics of delivery ratio, average delay and average number of trials per successfull transmission. / Espera-se que as redes de sensores e atuadores sem fio cresçam rapidamente em tamanho devido à contínua redução de custo dos dispositivos. Para tais redes de larga escala, os algoritmos de roteamento geográfico são adequados. Estes encontram a rota com base nas posições dos nós. Contudo, geralmente as rotas encontradas são mais longas do que as rotas possíveis, aumentando a probabilidade de perda do pacote. Ademais, os dispositivos sem fio de baixo custo operam nas faixas livres do espectro e estão sujeitos a interferências causadas por outras redes que operam nessas mesmas faixas. Ainda, com o aumento da densidade da rede, aumenta também a competição pelo canal. Portanto, tanto interferências internas quanto externas prejudicam a confiabilidade da comunicacão. Para aumentá-la pode-se utilizar múltiplos canais, uma vez que os rádios que equipam os nós podem ser sintonizados em diferentes frequências. Contudo, o número de canais disponíveis é limitado, tornando imperativo um protocolo de alocação eficiente. Porém foi demonstrado que encontrar uma solução ótima é inviável. Por isso, a Teoria dos Jogos pode ser empregada para se encontrar alocações sub-ótimas de forma distribuída. A Teoria dos Jogos permite modelar o problema como um jogo, onde os nós são os jogadores; os canais, suas ações; e conforme suas escolhas, cada não recebe uma recompensa. Um projetista deve definir as funções de recompensa e um algoritmo de aprendizado, que permite ao jogador determinar a ação que maximiza sua recompensa através da observação do jogo. Se o jogo for potencial, o algoritmo pode levá-lo a um equilíbrio, conhecido por Equilíbrio de Nash (NE), após repetidas rodadas. Em um NE, todo jogador estão satisfeito e sua recompensa pode ser limitada por uma constante multiplicativa da máxima recompensa possível. Seguindo esta abordagem, um protocolo denominado GBCA foi proposto na literatura, o qual aloca canais com base na topologia da rede e nas rotas utilizadas. Contudo, o GBCA necessita conhecer as rotas antecipadamente e estas devem ser estáticas. Por outro lado, algoritmos geográficos determinam a rota sob demanda para cada pacote. Portanto, para tornar o GBCA adequado para redes com roteamento geográfico, propõe-se, nesta Tese, modificações ao GBCA. Utilizando-se o protocolo proposto, denominado GBCA-G, demonstra-se, por simulação, que é possível obter resultados melhores quanto à taxa de entrega de pacotes, ao atraso médio de entrega e ao número médio de transmissões efetuadas por transmissão bem-sucedida.