Mars represents one of the most important targets for space exploration in the next 10 to 30 years, particularly because of evidence of liquid water in the planet's past. Current environmental conditions dictate that any existing water reserves will be in the form of ice; finding and sampling these ice deposits would further the study of the planet's climate history, further the search for evidence of life, and facilitate in-situ resource utilization during future manned exploration missions. This thesis presents a suite of algorithms to help enable a robotic ice-prospecting mission to Mars. Starting from visual odometry---the estimation of a rover's motion using a stereo camera as the primary sensor---we develop the following extensions: (i) a coupled surface/subsurface modelling system that provides novel data products to scientists working remotely, (ii) an autonomous retrotraverse system that allows a rover to return to previously visited places along a route for sampling, or to return a sample to an ascent vehicle, and (iii) the extension of the appearance-based visual odometry pipeline to an actively illuminated light detection and ranging sensor that provides data similar to a stereo camera but is not reliant on consistent ambient lighting, thereby enabling appearance-based vision techniques to be used in environments that are not conducive to passive cameras, such as underground mines or permanently shadowed craters on the moon. All algorithms are evaluated on real data collected using our field robot at the University of Toronto Institute for Aerospace Studies, or at a planetary analogue site on Devon Island, in the Canadian High Arctic.

Mars represents one of the most important targets for space exploration in the next 10 to 30 years, particularly because of evidence of liquid water in the planet's past. Current environmental conditions dictate that any existing water reserves will be in the form of ice; finding and sampling these ice deposits would further the study of the planet's climate history, further the search for evidence of life, and facilitate in-situ resource utilization during future manned exploration missions. This thesis presents a suite of algorithms to help enable a robotic ice-prospecting mission to Mars. Starting from visual odometry---the estimation of a rover's motion using a stereo camera as the primary sensor---we develop the following extensions: (i) a coupled surface/subsurface modelling system that provides novel data products to scientists working remotely, (ii) an autonomous retrotraverse system that allows a rover to return to previously visited places along a route for sampling, or to return a sample to an ascent vehicle, and (iii) the extension of the appearance-based visual odometry pipeline to an actively illuminated light detection and ranging sensor that provides data similar to a stereo camera but is not reliant on consistent ambient lighting, thereby enabling appearance-based vision techniques to be used in environments that are not conducive to passive cameras, such as underground mines or permanently shadowed craters on the moon. All algorithms are evaluated on real data collected using our field robot at the University of Toronto Institute for Aerospace Studies, or at a planetary analogue site on Devon Island, in the Canadian High Arctic.

The Space Flight Laboratory (SFL) at the University of Toronto Institute for Aerospace Studies is currently designing CanX-4 and CanX-5, a pair of formation-flying nanosatellites that will target centimeter-level position determination and sub-meter control. Once formation flight has been demonstrated, future missions can carry payloads designed to exploit these capabilities. Earth Observation is one such application that can benefit greatly from the availability of multiple platforms with precise position determination and attitude control. This work explores multistatic interferometric synthetic aperture radar (InSAR) as a particularly promising implementation of formation flight. Several mission scenarios are considered, including three commonly proposed InSAR constellation configurations, namely the Cartwheel, the Cross-Track Pendulum, and the Car-Pe configuration, as well as three large ( kilowatt) SAR transmitters (L-, C- and X-band) and one microsatellite transmitter (X-band, 150W). Using a framework of STK and MATLAB simulation and analysis tools, each case is evaluated with respect to the available interferometric baselines, ground coverage, resolution, and utility for selected applications including digital elevation modeling, moving target detection, and superresolution imagery. The “large” X-band transmitter is found to produce the most favorable operating area and resolution, and the Car-Pe configuration provides the greatest utility and flexibility for a combination of the three selected applications.

Ethylene/Air diffusion flames were studied at sub and super-atmospheric pressures to simulate a microgravity environment at fuel flow rates of 0.482 mg/s and 1.16 mg/s. Flame properties including flame dimensions, soot formation, temperature, and attachment mechanisms were investigated. Overall, luminous flame height decreased with decreasing pressure to the point of visible luminosity disappearance, resulting in blue flames. Flame width increased with decreasing pressure until the flame was almost spherical. Soot formation decreased with decreasing pressure to negligible concentrations in a near vacuum. At 0.482 mg/s, the percentage of carbon converted into soot was between 0.01% and 0.12%, whereas at 1.16 mg/s, this percentage was between 0.5% and 11% at sub-atmospheric pressures. Maximum flame temperatures increased with decreasing pressure. Regardless of fuel flow rate, the diffusion flames remained attached to the exterior of the burner. This attachment point moved further down the burner exterior as pressure decreased until a near vacuum.

Loss-of-control resulting from airplane upset is a leading cause of worldwide commercial aircraft accidents. One of the upset prevention and recovery strategies currently being considered is to provide pilot upset recovery training using ground-based flight simulators. However, to simulate the large amplitude and highly dynamic motions seen in upset conditions, both the flight model and the simulator motion need improvement. In this thesis, an enhanced flight model is developed to better represent the aircraft dynamics in upset conditions. In particular, extension is made to the aerodynamic database of an existing Boeing 747-100 (B-747) model to cover large angle of attack, sideslip and angular rates. The enhanced B-747 model is then used to conduct a set of upset recovery experiments in a flight simulator without motion. The experimental results can be used to identify and potentially correct major motion cueing errors caused by the conventional motion drive algorithm in upset conditions.

As international growth in wind energy steadily increases and the world gradually moves away from fossil fuels, advanced computational tools are required to produce accurate and fast predictions in wind turbine performance, and to allow efficient design cycles using advanced materials and manufacturing methods. Currently, aerostructural analysis often employs the relatively fast but inaccurate Blade Element Momentum (BEM) theory, while accurate but slower Computational Fluid Dynamics (CFD) methods are generally used for aerodynamic analysis alone.To bridge the gap between speed and accuracy, a 3D panel code, TriPan, was coupled with an advanced structural Finite Element Method (FEM) code, TACS, to perform aerostructural analysis for wind turbine blades. In addition, the framework allows the replacement of the panel solver by higher fidelity solvers to increase the accuracy of the overall aerostructural solution.

The problem of stand-off tracking of a moving target using a quadrotor unmanned aerial vehicle (UAV) based on vision-sensing is investigated. A PID (Proportional-Integral-Derivative) controller is implemented for attitude stabilization of the quadrotor. An LQG-based (Linear-Quadratic-Gaussian) control law is designed and implemented for position control of the quadrotor for a moving target tracking task. A novel vision-based estimation algorithm is developed, enabling estimation of quadrotor’s position, altitude and yaw relative to the target based on limited information about the target. Two image processing algorithms are implemented and compared for the task of feature detection and feature tracking in a series of images. Image processing algorithms are integrated with quadrotor control and experiments are performed to validate proposed control and estimation approaches.

The effect of end boundary conditions on the three-dimensionality of the vortex shedding from a circular cylinder in sub-critical flow has been studied experimentally, with a focus on the unsteady nature of the vortex filaments. Analysis of the near-wake of the cylinder was undertaken to determine the dependency of the spanwise uniformity of the vortex shedding on the end conditions. Flow visualization was performed downstream of the cylinder, and the temporal variation of the vortex filament angle was observed. Vortex dislocations were found to occur in this Reynolds Number regime regardless of the end boundary conditions. Having a cylinder bounded by two elliptical leading edge geometry endplates at an L/D value of five yielded parallel shedding with a reduction in the time-based variation of the vortex filament angle, and was shown to be the ideal end conditions for modeling an infinite cylinder in a free-surface water channel.

Loss-of-control resulting from airplane upset is a leading cause of worldwide commercial aircraft accidents. One of the upset prevention and recovery strategies currently being considered is to provide pilot upset recovery training using ground-based flight simulators. However, to simulate the large amplitude and highly dynamic motions seen in upset conditions, both the flight model and the simulator motion need improvement. In this thesis, an enhanced flight model is developed to better represent the aircraft dynamics in upset conditions. In particular, extension is made to the aerodynamic database of an existing Boeing 747-100 (B-747) model to cover large angle of attack, sideslip and angular rates. The enhanced B-747 model is then used to conduct a set of upset recovery experiments in a flight simulator without motion. The experimental results can be used to identify and potentially correct major motion cueing errors caused by the conventional motion drive algorithm in upset conditions.

As international growth in wind energy steadily increases and the world gradually moves away from fossil fuels, advanced computational tools are required to produce accurate and fast predictions in wind turbine performance, and to allow efficient design cycles using advanced materials and manufacturing methods. Currently, aerostructural analysis often employs the relatively fast but inaccurate Blade Element Momentum (BEM) theory, while accurate but slower Computational Fluid Dynamics (CFD) methods are generally used for aerodynamic analysis alone.To bridge the gap between speed and accuracy, a 3D panel code, TriPan, was coupled with an advanced structural Finite Element Method (FEM) code, TACS, to perform aerostructural analysis for wind turbine blades. In addition, the framework allows the replacement of the panel solver by higher fidelity solvers to increase the accuracy of the overall aerostructural solution.