In this thesis, wall-resolved and wall-modeled large-eddy simulation (LES) have been employed to investigate turbulent flow and dispersion around a single and a group of wall-mounted bluff bodies which are partially and fully submerged in developing boundary layers, respectively. The dispersion is caused by a continuous release of a passive scalar from a ground-level point source located within the matrix of obstacles. The results have been validated through comparisons against the available experimental measurement data. Thorough physical analysis including investigation of the spatial evolution and temporal cascades of the kinetic and scalar energies, flow structures and their influences on dispersion of the concentration plume in the context of highly disturbed flows, and study of turbulence statistics for the flow and concentration fields have been performed to provide deeper insights into turbulent flow and dispersion in domains with complex geometries. An in-house code based on FORTRAN programming language, parallelized with MPI libraries has been developed, modified and optimized for conducting the simulations. The simulations have been conducted on public-domain supercomputers ofWest-Grid, specifically Orcinus and Grex, and also the local 256-core cluster system of the CFD LAB at the University of Manitoba.

As indicated by drilling and geophysical data, the Mississippi Valley Graben, the Rough Creek Graben, together with the Rome Trough of eastern Kentucky and West Virginia, are fault-bounded graben structures filled with as much as 27,000 feet of Early to Middle Cambrian sediments. Detailed regional mapping of Cambrian and younger strata within and surrounding these structures indicates that they formed contemporaneously. The proximity of these structures suggests they developed within the same regional stress fields and tectonic environments. These three structures are mechanically and kinematically connected, and formed part of a single continent-scale rift system produced during the breakup of Rodinia and the separation of Laurentia from Amazonia. Data including stratigraphic tops from 1,764 wells, interpretations of 106 seismic profiles, aeromagnetic and gravity survey analysis, and mapped surface geology and structures were used within this project. Seven stratigraphic packages resolvable in both geophysical well logs and reflection seismic profiles were mapped in the subsurface across parts of Kentucky, Ohio, Indiana, Illinois, Missouri, and Tennessee. These stratigraphic units were then analyzed through structure maps, isopachous maps, and across 12 regional well-based cross sections. Detailed analysis of thickness patterns of seven major stratigraphic packages was used to identify the locations and timing of major fault movements within the study area. The regional patterns of fault movements through time were used to investigate how the structures evolved in response to the tectonic episodes in southeastern Laurentia during the Cambrian through Devonian Periods. Active rifting of the Precambrian crystalline bedrock began by the Early Cambrian, and resulted in a thick deposit of Reelfoot Arkose and Eau Claire Formation within the Mississippi Valley and Rough Creek Grabens, and the Rome Formation and Conasauga Group within the Rome Trough. Major tectonic extension ended by the Late Cambrian, prior to the deposition of the Knox Supergroup. Counter-clockwise rotation of the regional sigma-1 stress field between the Middle Ordovician and Early Mississippian (Taconic through Acadian Orogenies) resulted in the reactivation of varying sets of preexisting faults through time. The locations, orientations, and timing of these active faults relate to the deep architecture of the rift system.

This thesis presents a rough set approach to reinforcement learning. This is made possible by considering behaviour patterns of learning agents in the context of approximation spaces. Rough set theory introduced by Zdzisław Pawlak in the early 1980s provides a ground for deriving pattern-based rewards within approximation spaces. Learning can be considered episodic. The framework provided by an approximation space makes it possible to derive pattern-based reference rewards at the end of each episode. Reference rewards provide a standard for reinforcement comparison as well as the actor-critic method of reinforcement learning. In addition, approximation spaces provide a basis for deriving episodic weights that provide a basis for a new form of off-policy Monte Carlo learning control method. A number of conventional and pattern-based reinforcement learning methods are investigated in this thesis. In addition, this thesis introduces two learning environments used to compare the algorithms. The first is a Monocular Vision System used to track a moving target. The second is an artificial ecosystem testbed that makes it possible to study swarm behaviour by collections of biologically-inspired bots. The simulated ecosystem has an ethological basis inspired by the work of Niko Tinbergen, who introduced in the 1960s methods of observing and explaining the behaviour of biological organisms that carry over into the study of the behaviour of interacting robotic devices that cooperate to survive and to carry out highly specialized tasks. Agent behaviour during each episode is recorded in a decision table called an ethogram, which records features such as states, proximate causes, responses (actions), action preferences, rewards and decisions (actions chosen and actions rejected). At all times an agent follows a policy that maps perceived states of the environment to actions. The goal of the learning algorithms is to find an optimal policy in a non-stationary environment. The results of the learning experiments with seven forms of reinforcement learning are given. The contribution of this thesis is a comprehensive introduction to a pattern-based evaluation of behaviour during reinforcement learning using approximation spaces.

The problem considered in this thesis is the development of an autonomous prototype robot capable of gathering sensory information from its environment allowing it to provide feedback on the condition of specific targets to aid in maintenance of hydro equipment. The context for the solution to this problem is based on the power grid environment operated by the local hydro utility. The intent is to monitor power line structures by travelling along skywire located at the top of towers, providing a view of everything beneath it including, for example, insulators, conductors, and towers. The contribution of this thesis is a novel robot design with the potential to prevent hazardous situations and the use of rough coverage feedback modified reinforcement learning algorithms to establish behaviours.

This dissertation describes the design and implementation of a system that has been designed to extract 3D information from pairs of 2D images. System input consists of two images taken by an ordinary digital camera. System output is a full 3D model extracted from 2D images. There are no assumptions about the positions of the cameras during the time when the images are being taken, but the scene must not undergo any modifications. The process of extracting 3D information from 2D images consists of three basic steps. First, point matching is performed. The main contribution of this step is the introduction of an approach to matching image segments in the context of an approximation space. The second step copes with the problem of estimating external camera parameters. The proposed solution to this problem uses 3D geometry rather than the fundamental matrix widely used in 2D to 3D conversion. In the proposed approach (DirectGS), the distances between reprojected rays for all image points are minimised. The contribution of the approach considered in this step is a definition of an optimal search space for solving the 2D to 3D conversion problem and introduction of an efficient algorithm that minimises reprojection error. In the third step, the problem of dense matching is considered. The contribution of this step is the introduction of a proposed approach to dense matching of 3D object structures that utilises the presence of points on lines in 3D space. The theory and experiments developed for this dissertation demonstrate the usefulness of the proposed system in the process of digitizing 3D information. The main advantage of the proposed approach is its low cost, simplicity in use for an untrained user and the high precision of reconstructed objects.

By using the children's own thoughts and stories this theses aim to investigate and illustrate the phenomenon of rough and tumble play and its existence among five year olds in preschool.     The formulations of the questions are: What do children consider as rough and tumble play? - and what characterise that? From without a perspective of rough and tumble play, which kinds of rough and tumble play do the children play in preschool? - and how do those characterise? What makes rough and tumble play possible and accepted for the children to play in preschool? In this thesis I have investigated the questions above by interviewing 11 five years olds about how they consider rough and tumble play in preschool. From without Birgitta Knutsdotter Olofsson's and Torben Hangaard Rasmussen's view about play and rough and tumble play I created a perspective of rough and tumble play, which I used for analyzing the empirics. With rough and tumble play I mean games or plays that includes wrestling, hitting, kicking, pushing, chasing, running or/and the use of weapons against other human beings or other visualized beings or/and human beings. The conclusions are that children has many different opinions of what rough and tumble play is. Depends on who you are asking you will get a different perspective. The conclusions also shows that rough and tumble play can exist, not just by its self, but also in interplay with other kinds of games and plays. The children are using different strategies to make the rough and tumble play fit the restrictions put on by the educators in the preschool.

This thesis presents a rough set approach to reinforcement learning. This is made possible by considering behaviour patterns of learning agents in the context of approximation spaces. Rough set theory introduced by Zdzisław Pawlak in the early 1980s provides a ground for deriving pattern-based rewards within approximation spaces. Learning can be considered episodic. The framework provided by an approximation space makes it possible to derive pattern-based reference rewards at the end of each episode. Reference rewards provide a standard for reinforcement comparison as well as the actor-critic method of reinforcement learning. In addition, approximation spaces provide a basis for deriving episodic weights that provide a basis for a new form of off-policy Monte Carlo learning control method. A number of conventional and pattern-based reinforcement learning methods are investigated in this thesis. In addition, this thesis introduces two learning environments used to compare the algorithms. The first is a Monocular Vision System used to track a moving target. The second is an artificial ecosystem testbed that makes it possible to study swarm behaviour by collections of biologically-inspired bots. The simulated ecosystem has an ethological basis inspired by the work of Niko Tinbergen, who introduced in the 1960s methods of observing and explaining the behaviour of biological organisms that carry over into the study of the behaviour of interacting robotic devices that cooperate to survive and to carry out highly specialized tasks. Agent behaviour during each episode is recorded in a decision table called an ethogram, which records features such as states, proximate causes, responses (actions), action preferences, rewards and decisions (actions chosen and actions rejected). At all times an agent follows a policy that maps perceived states of the environment to actions. The goal of the learning algorithms is to find an optimal policy in a non-stationary environment. The results of the learning experiments with seven forms of reinforcement learning are given. The contribution of this thesis is a comprehensive introduction to a pattern-based evaluation of behaviour during reinforcement learning using approximation spaces.

The objective of this study is to determine the effect of different roughness types and arrangements on hydraulic jump characteristics in a rectangular channel. Three different types of roughness were used along experiments. All of them had rectangular prism shapes and that were placed normal to the flow direction. To avoid cavitation, height of roughness elements were arranged according to level of the channel inlet, so that the crests of roughness elements would not be protruding into the flow. The effects of roughness type and arrangement on hydraulic jump properties, i.e. energy dissipation, length of the jump and tail water depth were investigated. These properties were compared with the available data in literature and with the properties of hydraulic jump occurred on smooth bed.

Greater knowledge of the rough-toothed dolphin, Steno bredanensis, is needed to effectively contribute to conservation and management efforts for this species. The primary purpose of this research was to describe ecological and biological parameters for S. bredanensis that will be useful in future assessments of population stress. Several approaches were used to study S. bredanensis, including investigations of free-ranging populations, dead specimens, and captive individuals. Free-ranging rough-toothed dolphins distributed near small oceanic island environments were found to be more commonly sighted in-shore than off-shore. In the Windward islands of French Polynesia, this species preferred water depths of 1000 to 2000m and a distance of 1.8 to 5.5 km from the barrier reef. Group sizes ofrough-toothed dolphins sighted in French Polynesia range between 1 and 35 individuals with a mean size of 12.1. Endocrinology data for S. bredanensis was established in captive healthy and stranded individuals. Ranges and means were provided for progesterone, testosterone, cortisol and thyroid hormones. Changes in thyroid hormone concentrations were reflective of health status and testosterone appeared to be suppressed in ill individuals. Reproduction in S. bredanensis was investigated by determining the size and age range that this species attains sexual and physical maturity. Female rough-toothed dolphins attain sexual maturity by 9 to 10 years of age and males between 5 and 10 years at a similar length of approximately 216 cm. Physical maturity is generally reached at an older age and larger size for both males and females. Ecologically healthy and unheahhy populations of S. bredanensis were described in this investigation and these fmding will be useful in assessing future threats to this species.

Greater knowledge of the rough-toothed dolphin, Steno bredanensis, is needed to effectively contribute to conservation and management efforts for this species. The primary purpose of this research was to describe ecological and biological parameters for S. bredanensis that will be useful in future assessments of population stress. Several approaches were used to study S. bredanensis, including investigations of free-ranging populations, dead specimens, and captive individuals. Free-ranging rough-toothed dolphins distributed near small oceanic island environments were found to be more commonly sighted in-shore than off-shore. In the Windward islands of French Polynesia, this species preferred water depths of 1000 to 2000m and a distance of 1.8 to 5.5 km from the barrier reef. Group sizes ofrough-toothed dolphins sighted in French Polynesia range between 1 and 35 individuals with a mean size of 12.1. Endocrinology data for S. bredanensis was established in captive healthy and stranded individuals. Ranges and means were provided for progesterone, testosterone, cortisol and thyroid hormones. Changes in thyroid hormone concentrations were reflective of health status and testosterone appeared to be suppressed in ill individuals. Reproduction in S. bredanensis was investigated by determining the size and age range that this species attains sexual and physical maturity. Female rough-toothed dolphins attain sexual maturity by 9 to 10 years of age and males between 5 and 10 years at a similar length of approximately 216 cm. Physical maturity is generally reached at an older age and larger size for both males and females. Ecologically healthy and unheahhy populations of S. bredanensis were described in this investigation and these fmding will be useful in assessing future threats to this species.