Este trabalho traz um estudo de veículos rodoviários articulados pesados durante o processo de freagem incluindo uma revisão bibliográfica sobre aspectos humanos, ambientais, legais, veiculares e operacionais relacionados com o desempenho, a estabilidade e a segurança veicular e uma descrição da técnica dos mapas de desempenho. Tal técnica é original e inovadora e foi desenvolvida através de modelo matemático e respectivo programa computacional para analisar o desempenho e a estabilidade de veículo-trator-semi-reboque durante a freagem em trecho retilíneo. Com esta técnica é possível obter o desempenho e a estabilidade direcional de veículos articulados pesados na freagem, obedecendo as restrições geométricas do veículo-trator-semi-reboque e de peso estabelecidas pelo fabricante e pela legislação, para todas as condições de carregamento e qualquer condição operacional. Os resultados teórico-experimentais que validaram o modelo matemático adotado também são mostrados neste trabalho. Este trabalho foi desenvolvido com o apolo da Mercedes-Benz do Brasil, resultando numa integração entre Empresa e Universidade que trouxe uma contribuição efetiva para o uso do veículo-trator-semi-reboque com maior eficiência, e segurança, unindo esforços da Empresa e da Universidade em prol da segurança veicular e do desenvolvimento tecnológico do Brasil. / This work describes a study of articulated heavy vehicles during braking process including the state-of-the-art of human, environmental, legal, vehicular and operational aspects related with performance, stability and vehicular safety and a description of performance charts technique. This technique is original and innovative and was developed with mathematical model and related computer program to analize the tractor-semi-trailer performance and stability during the braking process in a straight line path. With this technique is possible to obtain the performance and directional stability of articulated heavy vehicles during braking process, obeying the geometrical restrictions of the tractor-semi-trailer and the legal and maker\'s weight restrictions, for all states of loading and any operational conditions. The theoretical-experimental results that validated the mathematical model adopted are also shown in this work. This work was developed with the support of Mercedes-Benz of Brazil, resulting in a Company and University integration that brought an effective contribution for the more efficient and safe use of tractor-semi-trailer, enlarging the Company and University efforts to the benefit of vehicular safety and technological development in Brazil.

A adulteração de combustíveis é uma grande preocupação no Brasil. A agência reguladora nacional (ANP) detecta anualmente de 1 a 3% de adulterações nas amostras coletadas, o que é um índice alto considerando o tamanho do mercado brasileiro. As alternativas de adulteração são vastas e muito dinâmicas, por isso os arranjos de sensores baseados no conceito de seletividade global parecem os mais adequados para detectar falsificação de combustíveis. O conceito de seletividade global leva em conta a sensibilidade cruzada de sensores químicos não específicos e o uso de métodos de análise multivariada de dados para encontrar padrões para amostras de diferentes composições químicas. Os sensores químicos podem ser obtidos de uma variedade de materiais sensoativos, cujas respostas elétricas variam de acordo com as propriedades físico-químicas do meio em que se encontra. Os polímeros condutores são excelentes materiais sensoativos, pois sua condutividade elétrica é grandemente influenciada pelas condições ambientais e podem ser processados na forma de filmes finos através várias técnicas. No presente trabalho, filmes de poli(3-metiltiofeno) (PMTh) e poli(3-hexiltiofeno) são depositados por cronopotenciometria e cronoamperometria sobre microeletrodos interdigitados e são caracterizados por espectroscopia de impedância. Os dados são analisados por redes neurais artificiais do tipo multilayer perceptron e bons resultados são obtidos na detecção de adulteração de gasolina. O mesmo estudo também pode ser aplicado na detecção de adulteração de álcool etílico combustível com um desempenho um pouco pior. / Fuel adulteration is a major concern in Brazil. The local governmental agency detects from 1 to 3% of problematic samples yearly, which is a lot considering Brazils market size. The myriad of adulteration possibilities is vast and it is very dynamic, thus array of sensors based on global selectivity concept seems to be more suitable methodology to detect problems in fuel. The global selectivity concept encompasses the cross-sensitivity of non-specific chemical sensors and the use of multivariated data analysis methods as a way to provide fingerprints for samples of different chemical composition. The chemical sensors can employ different types of sensoactive materials, whose electrical responses are dependent on the physicochemical characteristics of the media they get in contact with. Conducting polymers (CP) are per excellence suitable sensoactive materials, since their electrical conductivity is highly influenced by the environmental conditions and they can be easily processed in the thin film form by different techniques. In the present work films of poly(3-methylthiophene) (PMTh) and poly(3-hexylthiophene) (PHTh) are deposited by chronopotenciometry and chronoamperometry onto interdigitated microelectrodes and characterized through Impedance Spectroscopy. This data was analyzed with Multilayer perceptron neural networks and a very good performance is found in gasoline adulteration detection. A less great performance was also achieved in the investigation vehicular ethanol adulteration.

O ABS (Antilock Braking System) é um sistema suplementar ao sistema de freio de serviço normal, que impede o travamento da roda em qualquer situação de frenagem, principalmente em frenagens de emergência, e com isto garante estabilidade e dirigibilidade ao veículo e reduz, na grande maioria de situações, a distância de parada em relação ao sistema de freio normal. Por sua vez, o travamento de roda na frenagem tem uma forte relação com acidentes de trânsito. Esta dissertação apresenta uma avaliação da eficiência do ABS na redução de acidentes de trânsito nas condições de uso cotidianas do veículo, baseando-se em uma característica da dinâmica do veículo: a eficiência de frenagem. Usualmente, esta avaliação da eficácia do ABS na prevenção de acidentes é feita através de estatísticas de dados de acidentes e de registros de veículos. O travamento da roda ocorre quando a demanda de aderência de veículo, que está diretamente relacionada com a desaceleração desejada, supera a aderência disponibilizada pelo conjunto pneu pavimento. Portanto através de curvas de probabilidade de disponibilidade de aderência, curvas de probabilidade de demanda de aderência e estatísticas sobre o uso do veículo, consegue-se estimar a probabilidade da ocorrência do travamento de roda. Este trabalho apresenta estimativas de acionamentos do freio para que ocorra um travamento, de distância percorrida para que ocorra um travamento e tempo decorrido para que ocorra um travamento. As estimativas são feitas para pistas de concreto seco e concreto molhado. Para esclarecer o funcionamento do sistema, simulou-se o sistema ABS em Matlab/Simulink. Para isto, desenvolveu-se um controlador de estados finitos, o qual foi implementado em um modelo de veículo completo. A simulação realizada foi uma frenagem em linha reta até a parada completa do veículo em uma pista homogênea de alta aderência. Neste estudo verificou-se que o ABS tem grande potencial para reduzir acidentes de trânsito, tanto em pista molhada quanto em pista seca. O ABS apresenta um ganho na segurança veicular mesmo em relação a um sistema de freio perfeitamente balanceado. Deve-se ressalvar que este resultado pode ser alterado pelo fator motorista, alteração esta causada pelo desconhecimento do funcionamento e utilização do sistema ou mesmo mau uso. / The ABS (Antilock Braking System) is a supplementary system to the regular service brake system, which prevents the wheel lock in any braking situation, mainly in emergency braking maneuvers, and it will guarantee stability and steerability of the vehicle and reduces, in most of the situations, the stopping distance in relation to the regular brake system. By the way, the wheel lock when braking has a strong relationship with traffic accidents. This thesis presents an assessment of the effciency of the ABS in the reduction of traffic accidents in daily conditions of use of the vehicle, based on a dynamic feature of the vehicle: the braking efficiency. Usually, this evaluation of the effectiveness of the ABS in the prevention of accidents is done through database statistics of accidents and records of vehicles. The wheel lock occurs when the friction demand of the vehicle, which is directly related to the desired deceleration, is larger than the friction provided by the interaction between the tire and the road surface. So through probability curves of availability of friction, probability curves of friction demand and statistics on the use of the vehicle, it is possible to estimate the probability of occurrence of wheel lock-ups. This study presents estimations of brake applications to cause a lock-up, estimations of distance travelled to cause a lock-up and estimations of time so that a wheel locks. The estimations are made for dry and wet concrete roads. To clarify the operation of the system, it was simulated the ABS system in Matlab/Simulink. In order to do this, it was developed a finite state controller, which was implemented in a full vehicle model. The simulation carried out was a straight line brake until the complete stop of the vehicle in a homogeneous high friction road. This study found out that the ABS has a great potential to reduce traffic accidents, both on wet road and on dry road. The ABS presents a gain in vehicles safety even in relation to a perfectly balanced brake system. It should be noted that this result can be affected by the driver factor, this change is caused by ignorance of the operation and use of the system or even misuse.

La communication véhiculaire permet le développement de nouvelles applications multicast émergentes telles que la gestion de la flotte et la distribution des Points d'Intérêt (POI). Ces deux catégories d'applications nécessitent une communication multicast de l'Internet vers les réseaux véhiculaires (VANET). Afin de mettre en place une communication multicast adaptée au contexte de la communication Internet-vers-réseaux véhiculaires, notre travail traite de deux aspects différents. Tout d'abord, l'accessibilité des véhicules en mouvement au service Internet et en deuxième lieu, la dissémination du message dans les VANET. Nous introduisons un schéma d'adressage multicast basé sur les coordonnées géographiques des véhicules qui leur permet de s'auto-configurer d'une façon dynamique sans aucun besoin d'échanger des messages de signalisation avec Internet. Nous proposons aussi une approche simplifiée de gestion de la mobilité des véhicules dans le cadre des architectures Mobile IP et Proxy Mobile IP. Le but de cette approche est d'optimiser l'échange des messages avec les entités responsables de la gestion de la mobilité dans Internet. Afin d'étudier les mécanismes de dissémination appropriés aux applications de gestion de flottes, nous nous proposons de revisiter les techniques de routage multicast traditionnelles basées sur une structure de diffusion en arbre. Pour cela, nous étudions leur application aux réseaux véhiculaires. Nous présentons une étude théorique portant sur la durée de vie des liens entre les véhicules en milieux urbains. Ensuite, en utilisant la simulation, nous étudions l'application de Multicast Adhoc On Demand Vector, MAODV et proposons Motion-MAODV, une version adaptée de MAODV qui a pour objectif d'établir des routes plus robustes Enfin, concernat la dissémination multicast géolocalisée dans les applications POI, nous proposons le protocole de routage Melody qui permet une diffusion geocast en milieu urbain. A partir de simulations, nous constatons que, comparé aux protocoles de géo-brodcasting dans les milieux urbain très denses, Melody assure plus de fiabilité et d'efficacité lors de l'acheminement des données vers les zones géographiques de destination. / Vehicular communications allow emerging new multicast applications such as fleet management and point of interest (POI). Both applications require Internet-to-vehicle multicasting. These approaches could not be applied to vehicular networks (VANET) due to their dynamic and distributed nature. In order to enable such multicasting, our work deals with two aspects. First, reachability of the moving vehicles to the multicast service and second, multicast message dissemination in VANET. We introduce first a self-configuring multicast addressing scheme that allows the vehicles to auto-configure a dynamic multicast address without a need to exchange signalling messages with the Internet. Second, we propose a simplified approach that extends Mobile IP and Proxy Mobile IP. This approach aims at optimizing message exchange between vehicles and entities responsible for managing their mobility in Internet. To study the dissemination mechanisms that are suitable for fleet management applications, we propose to revisit traditional multicast routing techniques that rely on a tree structure. For this purpose, we study their application to vehicular networks. In particular, as vehicular networks are known to have changing topology, we present a theoretical study of the link lifetime between vehicles in urban environments. Then, using simulations, we study the application of Multicast Adhoc On Demand Vector, MAODV. We propose then Motion-MAODV, an improved version of MAODV that aims at enhancing routes built by MAODV in vehicular networks and guarantee longer route lifetime. Finally, to enable geographic dissemination as required by POI applications, we propose a routing protocol Melody that provides a geocast dissemination in urban environments. Through simulations, Melody ensures more reliable and efficient packet delivery to a given geographic area compared to traditional geo-brodcasting schemes in highly dense scenarios.

With an increasing need for greater traffic safety, there is an increasing demand for means by which solutions to the traffic safety problem can be studied. The purpose of this thesis is to investigate the feasibility of using an autonomous indoor navigation system as a component in a demonstration system for studying cooperative vehicular scenarios. Our method involves developing and evaluating such a navigation system. Our navigation system uses a pre-existing localization system based on passive RFID, odometry and a particle filter. The localization system is used to estimate the robot pose, which is used to calculate a trajectory to the goal. A control system with a feedback loop is used to control the robot actuators and to drive the robot to the goal.   The results of our evaluation tests show that the system generally fulfills the performance requirements stated for the tests. There is however some uncertainty about the consistency of its performance. Results did not indicate that this was caused by the choice of localization techniques. The conclusion is that an autonomous navigation system using the aforementioned localization techniques is plausible for use in a demonstration system. However, we suggest that the system is further tested and evaluated before it is used with applications where accuracy is prioritized.

There are still many challenges in the field of VANETs that encouraged researchers to conduct further investigation in this field to meet these challenges. The issue pertaining to routing protocols such as delivering the warning messages to the vehicles facing Non-Line of Sight (NLOS) situations without causing the storm problem and channel contention, is regarded as a serious dilemma which is required to be tackled in VANET, especially in congested environments. This requires the designing of an efficient mechanism of routing protocol that can broadcast the warning messages from the emergency vehicles to the vehicles under NLOS, reducing the overhead and increasing the packet delivery ratio with a reduced time delay and channel utilisation. The main aim of this work is to develop the novel routing protocol for a high-density environment in VANET through utilisation of its high mobility features, aid of the sensors such as Global Positioning System (GPS) and Navigation System (NS). In this work, the cooperative approach has been used to develop the routing protocol called the Co-operative Volunteer Protocol (CVP), which uses volunteer vehicles to disseminate the warning message from the source to the target vehicle under NLOS issue; this also increases the packet delivery ratio, detection of NLOS and resolution of NLOS by delivering the warning message successfully to the vehicle under NLOS, thereby causing a direct impact on the reduction of collisions between vehicles in normal mode and emergency mode on the road near intersections or on highways. The cooperative approach adopted for warning message dissemination reduced the rebroadcast rate of messages, thereby decreasing significantly the storm issue and the channel contention. A novel architecture has been developed by utilising the concept of a Context-Aware System (CAS), which clarifies the OBU components and their interaction with each other in order to collect data and take the decisions based on the sensed circumstances. The proposed architecture has been divided into three main phases: sensing, processing and acting. The results obtained from the validation of the proposed CVP protocol using the simulator EstiNet under specific conditions and parameters showed that performance of the proposed protocol is better than that of the GRANT protocol with regard to several metrics such as packet delivery ratio, neighbourhood awareness, channel utilisation, overhead and latency. It is also successfully shown that the proposed CVP could detect the NLOS situation and solves it effectively and efficiently for both the intersection scenario in urban areas and the highway scenario.

The increasing complexity of vehicular embedded systems has encouraged researchers and practitioners to adopt model-driven engineering in the development of these systems. In particular, several modelling languages have been introduced for representing the vehicular software architecture and its quality attributes. Current trend in the automotive domain is to shift from single-core architectures to multi-core ones in the attempt of providing the computational power required from the next generation of vehicles, particularly autonomous ones. On the one hand, multi-core architectures introduce new real-time challenges in the development of these systems like core-interdependency. On the other hand, it is pivotal that modelling languages continue to be effective in representing the new vehicular architectures together with their novel concerns, to continue to benefit from model-based methodologies. In this thesis, we present a systematic mapping study focusing i) on the real-time challenges introduced by the adoption of multi-core architectures and ii) on the extent of the modelling support for the resolution of these challenge, in the automotive domain.

Au cours des dernières années, on a observé l'intérêt croissant envers l'accessibilité à l'information et, en particulier, envers des approches innovantes utilisant les services à distance accessibles depuis les appareils mobiles à travers le monde. Parallèlement, la communication des véhicules, utilisant des capteurs embarqués et des dispositifs de communication sans fil, a été introduite pour améliorer la sécurité routière et l'expérience de conduite à travers ce qui est communément appelé réseaux véhiculaires (VANET). L'accès sans fil à l’Internet à partir des véhicules a déclenché l'émergence de nouveaux services pouvant être disponibles à partir ceux-ci. Par ailleurs, une extension du paradigme des réseaux véhiculaires a été récemment promue à un autre niveau. Le nuage véhiculaire (Vehicular Cloud) (VC) est la convergence ultime entre le concept de l’infonuagique (cloud computing) et les réseaux véhiculaires dans le but de l’approvisionnement et la gestion des services. Avec cette approche, les véhicules peuvent être connectés au nuage, où une multitude de services sont disponibles, ou ils peuvent aussi être des fournisseurs de services. Cela est possible en raison de la variété des ressources disponibles dans les véhicules: informatique, bande passante, stockage et capteurs. Dans cette thèse, on propose des méthodes innovantes et efficaces pour permettre la délivrance de services par des véhicules dans le VC. Plusieurs schémas, notamment la formation de grappes ou nuages de véhicules, la planification de transmission, l'annulation des interférences et l'affectation des fréquences à l'aide de réseaux définis par logiciel (SDN), ont été développés et leurs performances ont été analysées. Les schémas de formation de grappes proposés sont DHCV (un algorithme de clustering D-hop distribué pour VANET) et DCEV (une formation de grappes distribuée pour VANET basée sur la mobilité relative de bout en bout). Ces schémas de regroupement sont utilisés pour former dynamiquement des nuages de véhicules. Les systèmes regroupent les véhicules dans des nuages qui ne se chevauchent pas et qui ont des tailles adaptées à leurs mobilités. Les VC sont créés de telle sorte que chaque véhicule soit au plus D sauts plus loin d'un coordonnateur de nuage. La planification de transmission proposée implémente un contrôle d'accès moyen basé sur la contention où les conditions physiques du canal sont entièrement analysées. Le système d'annulation d'interférence permet d'éliminer les interférences les plus importantes; cela améliore les performances de planification d’utilisation de la bande passante et le partage des ressources dans les nuages construits. Enfin, on a proposé une solution à l'aide de réseaux définis par logiciel, SDN, où différentes bandes de fréquences sont affectées aux différentes liens de transmission de chaque VC afin d’améliorer les performances du réseau. / Abstract : In recent years, we have observed a growing interest in information accessibility and especially innovative approaches for making distant services accessible from mobile devices across the world. In tandem with this growth of interest, there was the introduction of vehicular communication, also known as vehicular ad hoc networks (VANET), leveraging onboard sensors and wireless communication devices to enhance road safety and driving experience. Vehicles wireless accessibility to the internet has triggered the emergence of service packages that can be available to or from vehicles. Recently, an extension of the vehicular networks paradigm has been promoted to a new level. Vehicular cloud (VC) is the ultimate convergence between the cloud computing concept and vehicular networks for the purpose of service provisioning and management. Vehicles can get connected to the cloud, where a multitude of services are available to them. Also vehicles can offer services and act as service providers rather than service consumers. This is possible because of the variety of resources available in vehicles: computing, bandwidth, storage and sensors. In this thesis, we propose novel and efficient methods to enable vehicle service delivery in VC. Several schemes including cluster/cloud formation, transmission scheduling, interference cancellation, and frequency assignment using software defined networking (SDN) have been developed and their performances have been analysed. The proposed cluster formation schemes are DHCV (a distributed D-hop clustering algorithm for VANET) and DCEV (a distributed cluster formation for VANET based on end-to-end relative mobility). These clustering schemes are used to dynamically form vehicle clouds. The schemes group vehicles into non-overlapping clouds, which have adaptive sizes according to their mobility. VCs are created in such a way that each vehicle is at most D-hops away from a cloud coordinator. The proposed transmission scheduling implements a contention-free-based medium access control where physical conditions of the channel are fully analyzed. The interference cancellation scheme makes it possible to remove the strongest interferences; this improves the scheduling performance and resource sharing inside the constructed clouds. Finally, we proposed an SDN based vehicular cloud solution where different frequency bands are assigned to different transmission links to improve the network performance.

Intelligent Transportation Systems (ITS) are gaining momentum among researchers. ITS encompasses several technologies, including wireless communications, sensor networks, data and voice communication, real-time driving assistant systems, etc. These states of the art technologies are expected to pave the way for a plethora of vehicular network applications. In fact, recently we have witnessed a growing interest in Vehicular Networks from both the research community and industry. Several potential applications of Vehicular Networks are envisioned such as road safety and security, traffic monitoring and driving comfort, just to mention a few. It is critical that the existence of convenience or driving comfort services do not negatively affect the performance of safety services. In essence, the dissemination of safety services or the discovery of convenience applications requires the communication among service providers and service requesters through constrained bandwidth resources. Therefore, service discovery techniques for vehicular networks must efficiently use the available common resources. In this thesis, we focus on the design of bandwidth-efficient and scalable service discovery protocols for Vehicular Networks. Three types of service discovery architectures are introduced: infrastructure-less, infrastructure-based, and hybrid architectures. Our proposed algorithms are network layer based where service discovery messages are integrated into the routing messages for a lightweight discovery. Moreover, our protocols use the channel diversity for efficient service discovery. We describe our algorithms and discuss their implementation. Finally, we present the main results of the extensive set of simulation experiments that have been used in order to evaluate their performance.

A real-time Vehicle Make and Model Recognition (VMMR) system is a significant component of security applications in Intelligent Transportation Systems (ITS). A highly accurate and real-time VMMR system significantly reduces the overhead cost of resources otherwise required. In this thesis, we present a VMMR system that provides very high classification rates and is robust to challenges like low illumination, occlusions, partial and non-frontal views. These challenges are encountered in realistic environments and high security areas like parking lots and public spaces (e.g., malls, stadiums, and airports). The VMMR problem is a multi-class classification problem with a peculiar set of issues and challenges like multiplicity, inter- and intra-make ambiguity among various vehicles makes and models, which need to be solved in an efficient and reliable manner to achieve a highly robust VMMR system. To reliably overcome the ambiguity challenges, a global features representation approach based on the Bag-of-Features paradigm is proposed. We extract key features from different make-model classes in an optimized dictionary, through two different dictionary building strategies. We represent different samples from each class with respect to the learned dictionary. We also present two classification schemes based on multi-class Support Vector Machines (SVMs): (1) Single multi-class SVM and (2) Attribute Bagging-based Ensemble of multi-class SVMs. These classification schemes allow simultaneous learning of the differences between global representations of different classes and the similarities between different shapes or generations within a same make-model class, to further overcome the multiplicity challenges for real-time application. Extensive experiments conducted using our approaches yield superior results for images that were occluded, under low illumination, partial camera views, or even non-frontal views, available in a recently published real-world VMMR dataset. The approaches presented herewith provide a highly accurate VMMR system for real-time applications in realistic environments.