Proper Orthogonal Decomposition (POD), combined with the Method of Snapshots and Galerkin projection, is a popular method for the model order reduction of nonlinear PDEs. The POD requires the left singular vectors from the singular value decomposition (SVD) of an n-by-m "snapshot matrix" S, each column of which represents the computed state of the system at a given time. However, the direct computation of this decomposition can be computationally expensive, particularly for snapshot matrices that are too large to fit in memory. Hierarchical Approximate POD (HAPOD) (Himpe 2016) is a recent method for the approximate truncated SVD that requires only a single pass over S, is easily parallelizable, and can be computationally cheaper than direct SVD, all while guaranteeing the requested accuracy for the resulting basis. This method processes the columns of S in blocks based on a predefined rooted tree of processors, concatenating the outputs from each stage to form the inputs for the next. However, depending on the selected parameter values and the properties of S, the performance of HAPOD may be no better than that of direct SVD. In this work, we numerically explore the parameter values and snapshot matrix properties for which HAPOD is computationally advantageous over the full SVD and compare its performance to that of a parallelized incremental SVD method (Brand 2002, Brand 2003, and Arrighi2015). In particular, in addition to the two major processor tree structures detailed in the initial publication of HAPOD (Himpe2016), we explore the viability of a new structure designed with an MPI implementation in mind. / Master of Science

Due to increasingly difficult threats, current air defense missile systems are pushed to the limits of their performance capabilities. In order to defend against these more stressing threats, interceptor missiles require greater maneuverability, faster response time, and increased robustness to more severe environmental conditions. One of the most critical missile system elements is the flight control system, since its time constant is typically half of the total missile system time constant. Conventional autopilot design techniques have worked well in the past, but in order to satisfy future and more stringent design specifications, new design methods are necessary. Robust control techniques (in particular, H-Infinity Control and Mu-Synthesis) and their application to the design of missile autopilots are addressed in this thesis. In addition, conventional autopilot designs are performed as comparative benchmarks. This paper reviews the missile autopilot design problem and presents descriptions of the classical and H-Infinity/Mu design methods. Missile autopilot designs considering both rigid-body dynamics and elastic-body dynamics are presented. Comparisons of the design approaches and results are also discussed. The results show that the application of robust control techniques to the design of missile autopilots can improve the performance and stability robustness characteristics of the flight control system. / Master of Science

In this dissertation we discuss two problems. In the first part, we consider the problem of computing a few extreme eigenvalues of a symmetric definite generalized eigenvalue problem or a few extreme singular values of a large and sparse matrix. The standard method of choice of computing a few extreme eigenvalues of a large symmetric matrix is the Lanczos or the implicitly restarted Lanczos method. These methods usually employ a shift-and-invert transformation to accelerate the speed of convergence, which is not practical for truly large problems. With this in mind, Golub and Ye proposes an inverse-free preconditioned Krylov subspace method, which uses preconditioning instead of shift-and-invert to accelerate the convergence. To compute several eigenvalues, Wielandt is used in a straightforward manner. However, the Wielandt deflation alters the structure of the problem and may cause some difficulties in certain applications such as the singular value computations. So we first propose to consider a deflation by restriction method for the inverse-free Krylov subspace method. We generalize the original convergence theory for the inverse-free preconditioned Krylov subspace method to justify this deflation scheme. We next extend the inverse-free Krylov subspace method with deflation by restriction to the singular value problem. We consider preconditioning based on robust incomplete factorization to accelerate the convergence. Numerical examples are provided to demonstrate efficiency and robustness of the new algorithm. In the second part of this thesis, we consider the so-called subspace clustering problem, which aims for extracting a multi-subspace structure from a collection of points lying in a high-dimensional space. Recently, methods based on self expressiveness property (SEP) such as Sparse Subspace Clustering and Low Rank Representations have been shown to enjoy superior performances than other methods. However, methods with SEP may result in representations that are not amenable to clustering through graph partitioning. We propose a method where the points are expressed in terms of an orthonormal basis. The orthonormal basis is optimally chosen in the sense that the representation of all points is sparsest. Numerical results are given to illustrate the effectiveness and efficiency of this method.

Birds are a key environmental asset and this is recognised through comprehensive legislation and policy ensuring their protection and conservation. Many species are active at night and surveys are required to understand the implications of proposed developments such as towers and reduce possible conflicts with these structures. Night vision devices are commonly used in nocturnal surveys, either to scope an area for bird numbers and activity, or in remotely sensing an area to determine potential risk. This thesis explores some practical and theoretical approaches that can improve the accuracy, confidence and efficiency of nocturnal bird surveillance. As image intensifiers and thermal imagers have operational differences, each device has associated strengths and limitations. Empirical work established that image intensifiers are best used for species identification of birds against the ground or vegetation. Thermal imagers perform best in detection tasks and monitoring bird airspace usage. The typically used approach of viewing bird survey video from remote sensing in its entirety is a slow, inaccurate and inefficient approach. Accuracy can be significantly improved by viewing the survey video at half the playback speed. Motion detection efficiency and accuracy can be greatly improved through the use of adaptive background subtraction and cumulative image differencing. An experienced ornithologist uses bird flight style and wing oscillations to identify bird species. Changes in wing oscillations can be represented in a single inter-frame similarity matrix through area-based differencing. Bird species classification can then be automated using singular value decomposition to reduce the matrices to one-dimensional vectors for training a feed-forward neural network.

The world is slowly moving into increased human-robot interaction where both humans and robots can co-exist in the same domain. For the robot to be able to operate effectively in a man’s designed environment, it becomes necessary to model the robot with human capabilities as humans are seen as more capable. Replicating human becomes a huge challenge due to numerous degrees-of-freedom (DOFs) that human possess resulting into too many variables and nonlinear equations. Other challenges do occur like singularities.   In this thesis, the singularity challenge of a redundant humanoid arm is explored while maintaining a simple 7 DOF serial chain structure. As opposed to the 30 DOF human arm, a simpler 7 DOF humanoid arm is adopted and studied to eliminate the singularity challenges. The singularity problem mainly comes from the elbow and the spherical joints at the shoulder and wrist. A step-by-step review of available inverse kinematics techniques is made with more focus on the iterative Jacobian-based methods. A step-by-step approach is adopted so as to identify the source of singularities while using the iterative Jacobian-based techniques that are able to handle the nonlinearities of the equations.   The Singular Value Filtering (SVF) technique coupled with Selectively Damped Least Squares (SDLS) is employed. Without any restrictions to the stretch of the arm or end-effector pose, the method demonstrates, in conjunction with Euler angle singularity avoidance method, the elimination of singularity problems. This is achieved with no adjustment to kinematic model of the manipulator.

National security needs dictate the development of new radar systems capable of identifying and tracking exoatmospheric threats to aid our defense. These new radar systems feature reduced noise floors, electronic beam steering, and ultra-wide bandwidths, all of which facilitate threat discrimination. However, in order to identify missile attributes such as RF reflectivity, distance, and velocity, many existing processing algorithms rely upon narrow bandwidth assumptions that break down with increased signal bandwidth. We present a fresh investigation into these algorithms for removing bandwidth limitations and propose novel state-space and direct-data factoring formulations such as * the multidimensional extension to the Eigensystem Realization Algorithm, * employing state-space models in place of interpolation to obtain a form which admits a separation and isolation of solution components, * and side-stepping the joint diagonalization of state transition matrices, which commonly plagues methods like multidimensional ESPRIT. We then benchmark our approaches and relate the outcomes to the Cramer-Rao bound for the case of one and two adjacent reflectors to validate their conceptual design and identify those techniques that compare favorably to or improve upon existing practices.

Frequency estimation plays an important role in many digital signal processing applications. Many areas have benefited from the discovery of the Fast Fourier Transform (FFT) decades ago and from the relatively recent advances in modern spectral estimation techniques within the last few decades. As processor and programmable logic technologies advance, unconventional methods for rapid frequency estimation in white Gaussian noise should be considered for real time applications. In this thesis, a practical hardware implementation that combines two known frequency estimation techniques is presented, implemented, and characterized. The combined implementation, using the well known FFT and a less well known modern spectral analysis method known as the Direct State Space (DSS) algorithm, is used to demonstrate and promote application of modern spectral methods in various real time applications, including Electronic Counter Measure (ECM) techniques.

Ce travail se propose de caractériser thermiquement des composites à microstructures complexes. Il s’agit de développer des méthodes d’estimation permettant d’identifier les propriétés thermiques des différentes phases en présence, ainsi que celles associées à leurs interfaces, à partir de mesures issues de la thermographie infrarouge. Cette estimation paramétrique nécessite la connaissance au préalable de la structure géométrique de l’échantillon. Le premier objectif concerne donc l’identification de la structure de l’échantillon testé par la discrimination des différentes phases et interfaces. Une fois la structure de l’échantillon connue, le second objectif est l’identification des paramètres thermiques des différents constituants ainsi que ceux de leurs interfaces. On se propose d’exploiter deux tests spécifiques utilisant le même dispositif expérimental. Deux méthodes mathématiques différentes ont été développées et utilisées pour exploiter les mesures de champ issues du premier test et permettre de retrouver la microstructure de l’échantillon. La première est fondée sur la décomposition en valeurs singulières des données de températures recueillies. Il est montré que cette méthode permet d’obtenir des représentations de la microstructure de très bonne qualité à partir de mesures même fortement bruitées. La seconde méthode permet de raffiner les résultats obtenus à l’aide de la méthode précédente. Elle repose sur la résolution d’un problème d’optimisation sous contraintes en exploitant la technique dite Level-Set pour identifier les frontières des différents constituants de l’échantillon. L’étape d’identification des propriétés thermiques des constituants et des interfaces exploite les mesures de champs issues du second test expérimental. La méthode développée, la SVD-FT combine des techniques de décompositions en valeurs singulières avec desfonctions tests particulières pour dériver des estimateurs linéaires des propriétés recherchées.Cette méthode permet de limiter les effets du bruit de mesure sur la qualité de l’estimation et de s’affranchir des opérations de filtrage des données. / This work reports on the thermal characterization of composites with a complex microstructure. It aims at developping mathematical methods to identify the thermal properties of the constituants and thoses associate at their interfaces. The first step consistsin discriminating the microstructure of the sample to be tested. Then, when the sample structure is known, the second step consists in estimating the thermal parameters of the different phases and those at their interfaces. One experimental device has been set up to realize those two steps. Two mathematical methods have been developped and used to discriminate the microstructure based on the images of the sample recorded bu an infrared camera. The first method is based on the singular value decomposition of the temperature data. It has been shown that this method gives a very good representation of the microstructure even with very noisy data. The second method allows to refine the results obtained by the first one. This method is based on the resolution of an optimization problem under constraints and use a Level-Set technic to identify the boundary of each phase. To estimate the thermal properties of each phase and its interface, the infrared images of the second experiment have been used. The SVD-FT method developed in this work combines the singular values decomposition technic with particular tests functions to derive linear estimat or for the thermal properties. As a result, a significant amplification of the signal/noise ratios is reached.

While magnetohydrodynamic (MHD) instabilities are considered one of the intriguing topics in tokamak physics, a feasibility study was conducted in the Saskatchewan Torus-Modified (STOR-M) tokamak to investigate the global MHD activities during the normal (L-mode) and improved (H-mode) confinement regimes. The experimental setup consists of 32 discrete Mirnov coils arranged into four poloidal arrays and mounted on STOR-M at even toroidal distances. The perturbed magnetic field fluctuations during STOR-M discharges were acquired and processed by the Fourier transform (FT), the wavelet analysis and the singular value decomposition (SVD) techniques. In L-mode discharges, the poloidal MHD mode numbers varied from 2 to 4 with peak frequencies in the range 20-40 kHz. The dominant toroidal modes were reported between 1 and 2 oscillating at frequencies 15-35 kHz. In another experiment, a noticeable MHD suppression was observed during the H-mode-like phase induced by the compact torus (CT) injection into STOR-M. However, a burst-like mode called the gong mode was triggered prior to the H-L transition, followed by coherent Mirnov oscillations. Mirnov oscillations with strong amplitude modulations were observed in the STOR-M tokamak. Correlations between Mirnov signals and soft x-ray (SXR) signals were found.

While magnetohydrodynamic (MHD) instabilities are considered one of the intriguing topics in tokamak physics, a feasibility study was conducted in the Saskatchewan Torus-Modified (STOR-M) tokamak to investigate the global MHD activities during the normal (L-mode) and improved (H-mode) confinement regimes. The experimental setup consists of 32 discrete Mirnov coils arranged into four poloidal arrays and mounted on STOR-M at even toroidal distances. The perturbed magnetic field fluctuations during STOR-M discharges were acquired and processed by the Fourier transform (FT), the wavelet analysis and the singular value decomposition (SVD) techniques. In L-mode discharges, the poloidal MHD mode numbers varied from 2 to 4 with peak frequencies in the range 20-40 kHz. The dominant toroidal modes were reported between 1 and 2 oscillating at frequencies 15-35 kHz. In another experiment, a noticeable MHD suppression was observed during the H-mode-like phase induced by the compact torus (CT) injection into STOR-M. However, a burst-like mode called the gong mode was triggered prior to the H-L transition, followed by coherent Mirnov oscillations. Mirnov oscillations with strong amplitude modulations were observed in the STOR-M tokamak. Correlations between Mirnov signals and soft x-ray (SXR) signals were found.