Le travail présenté dans cette thèse s'inscrit dans la problématique du transport éolien du sable.<br />L'étude principale porte sur un aspect particulier du transport par saltation. La saltation désigne le mouvement par saut des grains de sable emportés par le vent. Lors du rebond du grain en saltation sur le lit de sable, certains grains peuvent ètre éjectés. Ce mécanisme d'éjection a été étudié à partir d'une expérience modèle de collision d'une bille sur un empilement bidimensionnel de mêmes billes. Une étude détaillée de ce processus a été réalisée, en fonction de la vitesse d'impact de la bille incidente, de la hauteur de l'empilement, des propriétés élastiques des billes ainsi que de la géométrie de l'empilement. Des lois de distribution des vitesses d'éjection ont été établies. Par ailleurs, nous avons montré que le processus d'éjection est extrèmement sensible au désordre au sein de l'empilement: dans le cas d'un empilement désordonné, un plus grand nombre de billes est éjecté. Enfin, nous avons développé un modèle phénoménologique basé sur une description simplifiée de la propagation à travers l'empilement, de l'onde de déformation générée par le choc. Ce modèle permet d'interpréter certains résultats expérimentaux du processus d'éjection. <br />La dernière partie de cette thèse aborde certains aspects du transport éolien à une autre échelle: celle des structures éoliennes telles que les rides et les dunes. Un modèle de formation des rides éoliennes est proposé qui est capable de prédire l'évolution spatio-temporelle de ces structures. Les paramètres du modèle sont reliés directement aux lois d'éjection déterminées expérimentalement. Nous avons mis en évidence notamment un phénomène de murissement des rides: la longueur d'onde augmente au cours du temps, ce qui est observé expérimentalement.

La synthèse de lois de commande efficaces apparaît comme un enjeu important dans la réalisation autonome de tâches de navigation robotiques. Ce problème peut être abordé par différentes approches. L'une d'entre elles, la commande référencée capteur, permet de définir les boucles de commande directement à partir des mesures sensorielles au lieu de les exprimer en fonction de l'état du robot. Dans ce contexte, du fait de la richesse du signal vidéo, la vision apparaît comme un capteur privilégié pour la réalisation de tâches très variées de manière précise. Cependant, la commande référencée vision reposant sur la seule régulation des indices visuels dans l'image s'avère mal adaptée pour réaliser des tâches de navigation complexes dans des environnements encombrés d'obstacles. En effet, dans ce cas, il est nécessaire de garantir d'une part la sécurité du robot, et d'autre part la disponibilité permanente des indices visuels dans l'image. Ce sont précisément ces problèmes que nous avons voulu étudier dans le cadre de cette thèse. Notre contribution a consisté à définir des stratégies de commande référencées multi-capteurs pour un robot mobile réalisant une tâche référencée vision dans un environnement encombré d'obstacles susceptibles d'occulter le motif visuel. Nous avons tout d'abord proposé des lois de commandes permettant d'éviter à la fois les occultations et les collisions. Toutefois, les résultats obtenus ont montré que chercher à éviter simultanément ces deux phénomènes surcontraignait le mouvement du robot, limitant la gamme des missions réalisables. Nous avons alors développé une seconde approche consistant à tolérer temporairement la perte du signal visuel. Celle-ci repose sur l'exploitation de la réversibilité du lien vision/mouvement exprimé par le torseur d'interaction. Nous avons ainsi proposé dans un cadre général plusieurs méthodes (analytiques et numériques) de reconstruction du signal visuel lorsqu'il devient indisponible. Nous avons ensuite validé ch acune de ces méthodes dans le cas de la réalisation d'une tâche de navigation référencée vision dans un milieu encombré d'obstacles. Nous avons également montré l'intérêt de nos approches lorsque la caméra présente un défaut de fonctionnement pendant l'exécution d'une mission.

L'observation et l'étude des nuages moléculaires du milieu interstellaire nécessitent la connaissance <br />de données physiques fondamentales qui permettront d'avoir accès aux caractéristiques physiques de ce milieu, notamment les températures et les densités.<br />Ces régions ne sont généralement pas à l'équilibre thermodynamique et la détermination des populations des niveaux rotationnels, qui sont liées aux caractéristiques physiques du milieu, nécessite la connaissance des taux de tous les processus de peuplement radiatifs et collisionnels. De nombreux travaux ont été réalisés sur les coefficients d'Einstein et les taux de collisions d'un certain nombre de molécules. Pour les processus collisionnels, les perturbateurs principaux du milieu interstellaire sont H2, H et He.<br />De nouveaux instruments, permettant d'obtenir des résolutions spatiales et spectrales excellentes, verront bientôt le jour. Il s'agit de l'observatoire spatiale HERSCHEL et du futur interféromètre millimètrique ALMA. L'interprétation des futures observations réalisées grâce à ces instruments nécessite de connaître avec suffisamment de précision les taux de collision des molécules, voire de déterminer pour la première fois ces taux.<br />L'ammoniac NH3 étant un des principaux outils d'étude du milieu interstellaire moléculaire, nous nous sommes intéressés à la détermination des taux de collision avec He et H2 de cette molécule. Nous avons bénéficié de l'existence de surfaces de potentiel intermoléculaire récentes et plus précises qu'auparavant pour les systèmes NH3-He et NH3-H2. A l'aide du code de collision moléculaire MOLSCAT, nous avons pu déterminer les sections efficaces intégrales de collision de ces systèmes. De ces sections efficaces, il est possible d'en déduire les taux de collision des systèmes étudiés. <br />Les isotopomères deutérés de l'ammoniac NH2D, ND2H et ND3 ont, tous les trois, été détectés dans le milieu moléculaire. Cette découverte a eu lieu récemment pour les deux derniers (2000 et 2003). Jusqu'à aujourdh'ui, aucune valeur de taux de collision avec He de ces espèces deutérées n'avait été calculée. Les espèces deutérées ayant un grand intérêt dans l'étude de la chimie interstellaire et de son évolution, nous avons déterminé les premiers taux de collision de ces espèces avec l'hélium et avons ainsi publié les premiers résultats sur ce sujet.

The collision probability approach to neutron transport can be used to obtain the energy-dependent neutron spectrum in nuclear reactor systems as well as other quantities of interest. This method makes the approximation that the neutron distribution is constant within homogeneous regions, or cells, in the system. This assumption restricts geometries that can be modeled by the collision probability approach. The geometry modeled is typically an infinite lattice of two homogeneous cells: a fuel pin cylinder and the coolant that surrounds it. The transport of neutrons between the homogeneous cells is done using probabilities describing the chance that a neutron having a collision in one cell has its next collision in another cell. These collision probabilities can be cast in terms of escape and transmission probabilities for each cell. Some methods exist that extend the collision probability approach to systems composed of more than two homogeneous cells. In this work, we present a novel collision probability method, based on previous work by Schneider et al. (2006a), for an arbitrary number of cells. The method operates by averaging the transmission probabilities across cells of the same shape, and thus assumes a certain level of homogeneity across all cells. When using multigroup cross sections, which the collision probability approach requires, it is necessary to consider the effect that a system's geometry and composition has on those multigroup cross sections. The cross sections must be computed in a way that accounts for the resonance self-shielding that may reduce the reaction rates in the resonance region. The process of developing self-shielded cross sections in a heterogeneous system utilizes an escape cross section. We compute this escape cross section using the same collision probabilities used to obtain the energy spectrum. Results are presented for simple two-cell systems, and preliminary results for four-cell simulations are also given. An extension to the method is provided that accounts for the fact that in thermal systems the assumption of homogeneity is not always valid. / text

Human-robot interaction (HRI) is the future of robotics. It is essential in the expanding markets, such as surgical, medical, and therapy robots. However, existing industrial systems can also benefit from safe and effective HRI. Many robots are now being fitted with joint torque sensors to enable effective human-robot collision detection. Many existing and off-the-shelf industrial robotic systems are not equipped with these sensors. This work presents and demonstrates a method for effective collision detection on a system with motor current feedback instead of joint torque sensors. The effectiveness of this system is also evaluated by simulating collisions with human hands and arms. Joint torques are estimated from the input motor currents. The joint friction and hysteresis losses are estimated for each joint of an SIA5D 7 Degree of Freedom (DOF) manipulator. The estimated joint torques are validated by comparing to joint torques predicted by the recursive application of Newton-Euler equations. During a pick and place motion, the estimation error in joint 2 is less than 10 Newton meters. Acceleration increased the estimation uncertainty resulting in estimation errors of 20 Newton meters over the entire workspace. When the manipulator makes contact with the environment or a human, the same technique can be used to estimate contact torques from motor current. Current-estimated contact torque is validated against the calculated torque due to a measured force. The error in contact force is less than 10 Newtons. Collision detection is demonstrated on the SIA5D using estimated joint torques. The effectiveness of the collision detection is explored through simulated collisions with the human hands and arms. Simulated collisions are performed both for a typical pick and place motion as well as trajectories that transverse the entire workspace. The simulated forces and pressures are compared to acceptable maximums for human hands and arms. During pick and place motions with vertical and lateral end effector motions at 10mm/s and 25mm/s, the maximum forces and pressures remained below acceptable levels. At and near singular configurations some collisions can be difficult to detect. Fortunately, these configurations are generally avoided for kinematic reasons. / text

Collision of the Yakutat Terrane with North America in southern Alaska has driven growth of the Chugach-St. Elias orogen. Glaciation of the St. Elias Range has periodically increased since the Miocene, but began dominating erosion and spurred enhanced exhumation since the mid-Pleistocene transition at ~1 Ma. Ice associated with this glacial intensification carved cross-shelf sea valleys that connect the St. Elias Range to the deep-sea Surveyor Fan. A newly increased terrigenous sediment flux into the fan triggered the formation and growth of the Surveyor Channel. The change in geomorphology observed throughout Fan sequences allows us to characterize the influence that a glaciated orogen can have in shaping margin processes and the sediment pathways from source to sink. Seismic data also reveal an isolated, large, short runout, mass-transport deposit (MTD) buried in the Surveyor Fan. The MTD geometry, size and location on a convergent margin lend support to recent studies suggesting seismic strengthening and infrequent sediment failure on active margins. This study provides insight into the magnitude and scope of events required to cause submarine mega-slides and overcome higher than normal sediment shear strength, including the influence of climate and sea level change. Beneath the Surveyor Fan, integrated geophysical data reveals massive intraplate shearing, and a lack of oceanic crust magnetic lineaments in regions of Pacific Plate crust. We argue that stress from the Yakutat-North America collision transferred outboard to the Pacific Plate is the major driver for the deformation causing these features. This stress would have resulted in significant strain in the NE corner of the Pacific Plate, creating pathways for sill formation in the crust and Surveyor Fan. The collision further intensified as the thickest Yakutat portion began to subduct during the Pleistocene, possibly providing the impetus for the creation of the Gulf of Alaska Shear Zone, a >200 km zone of shear extending out into the Pacific Plate. This study highlights the importance of farfield stress from complex tectonic regimes in consideration of large-scale oceanic intraplate deformation. / text

An understanding of the atomic-scale features of chemical and physical processes taking place behind the shockwave front will help in addressing some of the major challenges in energetic materials research. The high pressure shockwave environment can be simulated using computational techniques to predict mechanical and chemical properties of a shocked material. Density functional theory calculations were performed to investigate uniaxial compressions of diamond and both hydrostatic and uniaxial compressions of TATB and NEST-1. For diamond, we calculated shear stresses for uniaxial compressions in the , , and directions and discovered the anomalous elastic regime which is responsible for the significant delay of plastic deformation behind a shockwave. For TATB, the hydrostatic equation of state, bulk modulus, and equilibrium structure were calculated using an empirical van der Waals correction. The principal stresses, shear stresses, and energy change per atom calculated for uniaxial compressions in the directions normal to the {001}, {010}, {011}, {100}, {101}, {110}, and {111} planes show highly anisotropic behavior. A similar study was performed for the newly synthesized energetic material NEST-1 in order to predict mechanical properties under uniaxial compression. From the similarities in the calculated principal stresses for each compression direction we conclude that NEST-1 is likely to exhibit relatively isotropic behavior as compared to other energetic materials. Finally, reactive molecular dynamics of shock-induced initiation chemistry in detonating PETN was investigated, using first-principles density functional theory, in order to identify the reaction mechanisms responsible for shock sensitivities in energetic materials. The threshold collision velocity of initiation for each orientation was determined and correlated with available experimental data on shock sensitivity. The production of NO2 was found to be the dominant reaction pathway in every reactive case. The simulations show that the reactive chemistry of initiation occurs at very short time scales ~10E?¹³ s at highly non-equilibrium conditions, and is driven by dynamics rather than temperature.

The research presented in this dissertation was performed in the Atomic, Molecular and Optical (AMO) physics laboratory at the University of Manitoba. Atomic beams of the two-valence-electron heavy atom systems, barium and ytterbium, were investigated with low energy electron scattering and optical emission studies. Both the ground states and laser excited states were investigated as a function of incident electron beam energy from 10 eV to 50 eV. Measurements of relative cross sections and polarization for 583 nm and 554 nm line emission from the (6s7p)1P1 and (6s6p)1P1 states of barium excited by electron impact from both the ground states and the optically pumped metastable (6s5d)1D2 are reported. Data are normalized to absolute cross sections for the ground state (6s2)1S0→(6s6p)1P1 state transition due to electron scattering, with corrections for branching ratios and cascading from higher states to deduce the total level excitation cross sections. Results are also presented for the first studies of the 399 nm line emission from laser-excited ytterbium, yielding an upper limit on the apparent cross section for the (6s6p)3P1→(6s6p)1P1 transition. Results are compared with the latest theoretical models and previous data, where available. / October 2015

In this dissertation, we study the evolution and properties of the relic (or cosmic) neutrino distribution from neutrino freeze-out at T=O(1) MeV through the free-streaming era up to today, focusing on the deviation of the neutrino spectrum from equilibrium and in particular we demonstrate the presence of chemical non-equilibrium that continues to the present day. The work naturally separates into two parts. The first focuses on aspects of the relic neutrinos that can be explored using conservation laws. The second part studies the neutrino distribution using the full general relativistic Boltzmann equation. Part one begins with an overview of the history of the Universe, from just prior to neutrino freeze-out up through the present day, placing the history of cosmic neutrino evolution in its proper context. Motivated by the Planck CMB measurements of the effective number of neutrinos, we derive those properties of neutrino freeze-out that depend only on conservation laws and are independent of the details of the scattering processes. Part one ends with a characterization of the present day neutrino spectrum as seen from Earth. The second part of this dissertation focuses on the properties of cosmic neutrinos that depend on the details of the neutrino reactions, as is necessary for modeling the non-thermal distortions from equilibrium and computing freeze-out temperatures. We first develop some geometry background concerning volume forms and integration on submanifolds that is helpful in computations. We then detail a new spectral method for solving the Boltzmann equation, based on a dynamical basis of orthogonal polynomials. Next, we detail an improved procedure for analytically simplifying the corresponding scattering integrals for subsequent numerical computation. Using this, along with the spectral method mentioned above, we solve the Boltzmann equation through the neutrino freeze-out period. Finally, we conclude by using our novel solution methods to perform parametric studies of the dependence of the neutrino freeze-out standard model parameters. This exploration is performed with the aim of recognizing mechanisms in the neutrino freeze-out process that are capable of leading to the measured value of the effective number of neutrinos.

This dissertation presents an investigation into the structure of macromolecular protein assemblies by mass spectrometry. The experiments described within are designed to systematically assess the analytical utility of surface-induced dissociation (SID) tandem mass spectrometry in the characterization of multi-subunit protein complexes. This is accomplished by studying the effects of ion-surface collision on the fragmentation products of protein assemblies that vary by mass, number of subunits, and protein structural features. The dissociation energetics and mechanisms of protein complexes are considered by examining the influence of ion internal energy and sub-oligomeric protein structure on the dissociation process. Conditions are first established for the preservation of “native” protein quaternary structure and applied to previously characterized systems for proof-ofconcept. These conditions are subsequently extended to determine the molecular weight and subunit stoichiometry of several small heat shock proteins. Native mass spectrometry is then combined with limited proteolysis experiments to characterize the subunit interface of a unique small heat shock protein, Hsp18.5 from Arabidopsis thaliana, identifying regions of the protein essential for preservation of the native dimer. The dissociation of non-covalent protein assemblies is then explored on a quadrupole time-of-flight (Q-TOF) mass spectrometer, modified for the study of ion-surface collisions. This instrument allows ions to be dissociated through collisions with a surface or more conventional collisions with gas atoms. The dissociation of protein complexes is explored by both activation methods beginning with specific and non-specific dimers with masses less than 40 kDa. These studies are extended to larger assemblies with as many as 14 subunits weighing over 800 kDa, and are applied to both homo- and hetero-oligomeric protein complexes. Activation of a protein complex with “n” subunits through multiple collisions with inert gas atoms results in asymmetric dissociation into a highly charged monomer and complementary (n-1)-mer regardless of protein size or subunit architecture. This process is known to occur through an unfolding of the ejected subunit, and limits the amount of structural insight that can be gleaned from such studies. Collision at a surface however, results in more charge and mass symmetric fragmentation, and in some instances reflects the substructure of the protein assembly under investigation. The differences observed between the CID and SID of protein complexes is attributed to the rapid deposition of large amounts of internal energy deposited upon collision at a more massive target such as a surface. The ion activation time-frame and energy transfer efficiency are proposed to induce dissociation on a time-scale that precedes subunit unfolding providing access to dissociation pathways that are inaccessible by traditional means of activation. The systems studied here represent the largest ions fragmented via surface collisions within a mass spectrometer, and the fragmentation products observed by SID demonstrate its promise for expanding the role of mass spectrometry in the field of structural biology.