Concerns over the environmental impacts of the transport sector have led to the United Kingdom (UK) Government establishing a legally binding commitment of an 80% reduction in greenhouse gas emissions by 2050 (relative to the 1990 baseline) through the Climate Change Act 2008. The decarbonisation of the transport sector by 2050 will substantially contribute towards achieving this target. Technological innovations, therefore, have an important role in supporting policy objectives. One innovation that is being developed for this purpose in the transport sector is an alternative fuel vehicle. While there are several alternative fuel vehicle technologies, the only two with zero tailpipe (exhaust) emissions are battery electric vehicles and hydrogen fuel cell vehicles. Both of these technologies are not yet at a stage in their development where they can successfully compete with conventional fuel vehicles (internal combustion engine vehicles). They face a variety of technological hurdles that include range, performance, cost, and infrastructure. Hydrogen fuel cell vehicles are not commercially available, although battery electric vehicles have been on the commercial market for several years. Uptake of alternative fuel vehicles is occurring at a slower pace than hoped by policy makers and manufacturers. The aim of this thesis is to examine the factors influencing the decision to adopt an alternative fuel vehicle, and is underpinned by Rogers (2003) Diffusion of Innovations theory. The Innovation-Decision Process from this theory posits that an individual must first know about an innovation before forming an attitude about it. Innovativeness is instrumental in determining the knowledge an individual has of an innovation and how early in the diffusion process they are likely to become an adopter. Perceptions of the innovation are influential in forming an attitude towards it. The focus of the research is on Birmingham, the UK s second largest city. The first stage of the research involves establishing the locations of individuals across the city that possess socio-demographic characteristics associated with early adopters of alternative fuel vehicles. This is achieved by applying cluster analysis to Birmingham census data, which enabled the identification of a strong spatial cluster of potential early adopters in the suburb of Sutton Coldfield. In the second stage of the research, a household questionnaire was undertaken with 413 respondents in Sutton Coldfield. The analysis of the questionnaire data firstly involves the verification of the early adopter characteristics from stage one by examining the relationship of these characteristics with innovativeness. Analysis is then undertaken of the level of knowledge and the perceptions that the respondents have of alternative fuel vehicles. The final step in the analysis is an evaluation of the characteristics of current models of electric vehicles and how well aligned they are with the driving needs and vehicle expectations of respondents. The results confirm that the knowledge of alternative fuel vehicles is limited and individual perceptions have led to the development of negative attitudes towards them. Socio-demographic characteristics were significant in influencing these factors. There were 5% (21) of respondents who have previously considered the adoption of an electric vehicle but have not yet done so. There is evidence from the survey of active rejection among a small number of respondents. The reasons largely relate to three problems: purchase price, limited range, and poor infrastructure availability. However, the majority of respondents have passively rejected alternative fuel vehicles, such that they have never given consideration to the adoption of one. This confirms that a concerted effort is required to inform the general public about alternative fuel vehicles. Opportunities for increasing adoption have been identified for policy and marketing, including education and awareness-raising campaigns.

In order for autonomous vehicles to achieve life-long operation in outdoor environments, navigation systems must be able to cope with visual change---whether it's short term, such as variable lighting or weather conditions, or long term, such as different seasons. As a GPS is not always reliable, autonomous vehicles must be self sufficient with onboard sensors. This thesis examines the problem of localisation against a known map across extreme lighting and weather conditions using only a stereo camera as the primary sensor. The method presented departs from traditional techniques that blindly apply out-of-the-box interest-point detectors to all images of all places. This naive approach fails to take into account any prior knowledge that exists about the environment in which the robot is operating. Furthermore, the point-feature approach often fails when there are dramatic appearance changes, as associating low-level features such as corners or edges is extremely difficult and sometimes not possible. By leveraging knowledge of prior appearance, this thesis presents an unsupervised method for learning a set of distinctive and stable (i.e., stable under appearance changes) feature detectors that are unique to a specific place in the environment. In other words, we learn place-dependent feature detectors that enable vastly superior performance in terms of robustness in exchange for a reduced, but tolerable metric precision. By folding in a method for masking distracting objects in dynamic environments and examining a simple model for external illuminates, such as the sun, this thesis presents a robust localisation system that is able to achieve metric estimates from night-today or summer-to-winter conditions. Results are presented from various locations in the UK, including the Begbroke Science Park, Woodstock, Oxford, and central London.

No description available.

Road traffic safety has been a subject of worldwide concern. Dedicated short range communications (DSRC) is widely regarded as a promising enabling technology for collaborative safety applications (CSA), which can provide robust communication and affordable performance to build large scale CSA system. The main focus of this thesis is to develop solutions for DSRC QoS control in order to provide robust QoS support for CSA. The first design objective is to ensure robust and reliable message delivery services for safety applications from the DSRC networks. As the spectrum resources allocated to DSRC network are expected to be shared by both safety and non-safety applications, the second design objective is to make QoS control schemes bandwidth-efficient in order to leave as much as possible bandwidth for non-safety applications. The first part of the thesis investigates QoS control in infrastructure based DSRC networks, where roadside access points (AP) are available to control QoS control at road intersections. After analyse DSRC network capabilities on QoS provisioning without congestion control, we propose a two-phases adaptive QoS control method for DSRC vehicle networks. In the first phase an offline simulation based approach is used to and out the best possible system configurations (e.g. message rate and transmit power) with given numbers of vehicles and QoS requirements. It is noted that with different utility functions the values of optimal parameters proposed by the two phases centralized QoS control scheme will be different. The conclusions obtained with the proposed scheme are dependent on the chosen utility functions. But the proposed two phases centralized QoS control scheme is general and is applicable to different utility functions. In the second phase, these configurations are used online by roadside AP adaptively according to dynamic traffic loads. The second part of the thesis is focused on distributed QoS control for DSRC networks. A framework of collaborative QoS control is proposed, following which we utilize the local channel busy time as the indicator of network congestion and adaptively adjust safety message rate by a modified additive increase and multiplicative decrease (AIMD) method in a distributed way. Numerical results demonstrate the effectiveness of the proposed QoS control schemes.

Cette thèse traite de la prise en compte de l'impact des procédés d'assemblage thermomécanique sur la géométrie de la caisse en blanc lors de la phase de conception d'un véhicule, chez PSA Peugeot-Citroën. Pour cela, on souhaite développer un outil de prévision des déformations induites par les procédés de soudo-brage - plasmatron et laser - lors de l'assemblage du pavillon sur le côté de la caisse. Les pièces - de dimensions métriques - sont composées de tôles fines d'épaisseur 0,67 mm en acier XES. Afin de réaliser une simulation numérique du procédé, il est nécessaire d'identifier un certain nombre de modèles, objets de cette thèse. Dans un premier temps des essais de soudo-brasage en laboratoire ont été réalisés afin de définir les domaines de validité de l'étude, d'identifier des modèles de sources de chaleur, et de déterminer les coefficients d'échange thermique avec l'extérieur. Une étude spécifique d'identification des lois de comportement des matériaux mis en jeu, sur toute la gamme de température, a également été réalisée. Pour cela, une méthode de mesure de champs par corrélation d'images numériques à haute température a été développée, qui a permis d'identifier complètement les paramètres de la loi de comportement de l'acier XES. Pour finir, de façon à réaliser la validation de la simulation numérique, ces travaux de thèse ont été l'occasion de développer une nouvelle méthode de calibration pour la stéréocorrélation d'images basée sur la connaissance a priori de la géométrie de l'objet observé, en considérant sa forme théorique via son modèle CAO. Cette méthode est tout particulièrement adaptée aux dimensions des structures étudiées ici. / This study is about the impact of thermo-mechanical assembly processes of metal sheets, in the automotive industry context. The aim is to predict thermally induced deformations by using a numerical tool. We are particularly interested in Plasmatron and laser brazing processes that are used to assemble an automotive roof and the body side of the vehicle. Parts are made from 0,67 mm thickness XES thin metal sheet of about one meter length. To validate such complex non-linear numerical simulations with experimental observations, it is necessary to develop well-controlled and highly instrumented tests. In a first step, experimental welding brazing tests are carried out in order to identify validity domains, heat source model and thermal coefficients of exchange. In a second step, high temperature tension tests are performed to identify the behaviour of each material. These tests are instrumented with a new DIC protocol in order to measure displacement fields from 20°C to 1000°C. Finally, in order to validate the thermo-mechanical simulation, in-situ 3D Digital Image Correlation (stereo-correlation) measurements were performed during the welding-brazing assembly. For this purpose and due to the geometry of the involved parts, a new calibration method, based on the CAD part geometry has been developed.

Les instruments politiques sur les émissions de véhicules passagers visent à réduire les externalités négatives sur l'environnement causées par l'usage des véhicules. Des réglementations sur les émissions de CO2 ont été mises en place en Europe, aux États-Unis, en Chine et ailleurs. La cible réglementaire basée sur la moyenne des émissions des véhicules vendus par un constructeur devient plus contraignante au fil du temps. Cette thèse analyse comment les constructeurs automobiles anticipent et préparent leurs futurs portefeuilles de technologies afin de respecter les futurs objectifs politiques. Pour conduire cette analyse, cette thèse développe un modèle d'optimisation des choix technologiques sous la contrainte de diffusion technologique.Avec ce cadre de modélisation basé sur la limitation de la vitesse à laquelle une technologie peut se diffuser dans un marché, cette thèse étudie trois questions politiques. Dans un premier temps, nous analysons comment le type d'anticipation du futur peut modifier les choix technologiques faits à court et à long termes. Nous montrons qu'une anticipation du futur focalisée sur les objectifs de court terme peut empêcher l'atteinte de la cible à long terme. Respecter la cible à court terme n'est une condition ni nécessaire ni suffisante pour permettre le niveau d'émissions requis par la cible à long terme. De plus si l'anticipation du futur n'est pas parfaite, les choix technologiques vont être verrouillés dans des technologies à faible potentiel d'abattement créant ainsi une dépendance au sentier qui limite l'abattement potentiel à long terme.Dans un deuxième temps, nous nous intéressons à évaluer quantitativement comment l'indexation sur la masse des véhicules de la réglementation CO2 change les critères optimaux de choix. Nous montrons qu'il n'existe pas de différence significative dans le coût social de la mobilité entre les deux mécanismes de réglementation CO2 avec et sans indexation sur la masse pour une même cible d'émissions. Cependant les choix technologiques entre ces mécanismes sont différents, la réglementation CO2 indexée à la masse ne développe en aucun cas les technologies d'allègement.Dans un troisième temps, nous étudions comment les choix technologiques changent quand des politiques à objectifs multiples se superposent. Nous centrons notre analyse sur deux externalités associées à la mobilité: les émissions CO2 et la pollution de l'air locale. Nous montrons trois types d'impacts de la superposition de politiques. Premièrement, une politique technologiquement spécifique tel que le Mandat de Véhicule à Zéro Émission en combinaison avec la réglementation CO2 provoque le développement de technologies vertes coûteuses et empêche les technologies sales et peu coûteuses de disparaître. Dans le cas de l'application de la réglementation CO2 seule nous n'observons pas ce comportement. Deuxièmement, la superposition de politiques peut mener à un coût élevé quand les technologies adaptées à chacune des politiques sont très différentes. Troisièmement, nous trouvons un effet ambigu de la superposition de politiques relative à l'application d'une politique seule sur la performance environnementale. / Policy instruments on passenger vehicle emissions aim at reducing negative environmental externalities from vehicles use. To regulate CO2 emissions, fuel economy standards have been put in place in Europe and in the US, among others. These standards are made more stringent over time. This thesis analyzes how automotive firms anticipate and prepare their future technology portfolio to comply with expected future standards. To do so, we develop a model of optimal technology choice that captures technology diffusion constraints.With this framework, this thesis investigates three policy questions. First, we ask how the form of anticipation can affect near- and long-term technology choices. We find that focusing solely on near-term objectives can lead to failure to comply with a long-term target. In fact, meeting the near-term target is not a necessary nor a sufficient condition to satisfy long-term compliance. Moreover, when there is partial anticipation, as in a myopic view of the future, technology choices will be stuck with low abatement technologies creating a path dependency that limits long-term abatement potential.Second, we ask how much indexing fuel economy standard to mass (as in Europe or China) changes the optimal technology. We show that, for the same emission target, there is no significant difference in the social cost of mobility for an average vehicle with and without mass index. Thus a heavier vehicle fleet has the same cost than a lighter one. However, the technology choices are different, and mass indexed fuel economy standards lead to sidestepping lightweight technologies despite being cost effective from a CO2 emissions abatement point of view.Third, we ask how technology choices change when policies with multiple objectives overlap. We focus on two externalities associated with mobility: CO2 emissions and local air pollution. We show three type of effects of overlapping policies. First, a technology specific policy such as the Zero Emission Vehicle Mandate in combination with a fuel economy standard induces carmakers to develop more expensive green technologies and prevents cheap, dirty technologies from disappearing compared to the case of a fuel economy standard alone. Second, the combination of policies can lead to very high costs when technologies adapted to each policy are very different. Third, we find an ambiguous effect of overlapping policies relative to single-objective policy in terms of emissions performance.