Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1993. / Vita. Abstract. Includes bibliographical references (leaves 59-60). Also available via the Internet.

Thesis (M.S. in Astronautical Engineering)--Naval Postgraduate School, December 2003. / "December 2003". Thesis advisor(s): Cristi, Roberto ; Agrawal, Brij. Includes bibliographical references (p. 53). Also available online.

Thesis (M. Eng.)(Electronic)--University of Pretoria, 2005. / Summaries in English and Afrikaans. Includes bibliographical references.

Thesis (M.S.)--University of Wisconsin--Madison. / Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 259-264).

Doctor of Philosophy / Department of Electrical and Computer Engineering / Balasubramaniam Natarajan / Stochastic hybrid system (SHS) is a class of dynamical systems that experience interaction of both discrete mode and continuous dynamics with uncertainty. State estimation for SHS has attracted research interests for decades with Kalman filter based solutions dominating the area. Mode-based Kalman filter is an extended version of the traditional Kalman filter for SHS. In general, as Kalman filter is unbiased for non-hybrid system estimation, prior research efforts primarily focus on the behavior of error covariance. In SHS state estimate, mode mismatch errors could result in a bias in the mode-based Kalman filter and have impacts on the continuous state estimation quality. The relationship between mode mismatch errors and estimation stability is an open problem that this dissertation attempts to address. Specifically, the probabilistic model of mode mismatch errors can be independent and identically distributed (i.i.d.), correlated across different modes and correlated across time. The proposed approach builds on the idea of modeling the bias evolution as a transformed system. The statistical convergence of the bias dynamics is then mapped to the stability of the transformed system. For each specific model of the mode mismatch error, the system matrix of the transformed system varies which results in challenges for the stability analysis. For the first time, the dissertation derives convergence conditions that provide tolerance regions for the mode mismatch error for three mode mismatch situations. The convergence conditions are derived based on generalized spectral radius theorem, Lyapunov theorem, Schur stability of a matrix polytope and interval matrix method. This research is fundamental in nature and its application is widespread. For example, the spatially and timely correlated mode mismatch errors can effectively capture cyber-attacks and communication link impairments in a cyber-physical system. Therefore, the theory and techniques developed in this dissertation can be used to analyze topology errors in any networked system such as smart grid, smart home, transportation, flight management system etc. The main results provide new insights on the fidelity in discrete state knowledge needed to maintain the performance of a mode-based Kalman filter and provide guidance on design of estimation strategies for SHS.

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Tecnologia, Departamento de Engenharia Mecânica, 2013. / Submitted by Albânia Cézar de Melo (albania@bce.unb.br) on 2013-08-14T12:57:54Z No. of bitstreams: 1 2013_SergioMessiasCruz.pdf: 7213334 bytes, checksum: 9b766d528b04c26ebdfe9d72d6924318 (MD5) / Approved for entry into archive by Guimaraes Jacqueline(jacqueline.guimaraes@bce.unb.br) on 2013-08-14T13:31:25Z (GMT) No. of bitstreams: 1 2013_SergioMessiasCruz.pdf: 7213334 bytes, checksum: 9b766d528b04c26ebdfe9d72d6924318 (MD5) / Made available in DSpace on 2013-08-14T13:31:25Z (GMT). No. of bitstreams: 1 2013_SergioMessiasCruz.pdf: 7213334 bytes, checksum: 9b766d528b04c26ebdfe9d72d6924318 (MD5) / Este trabalho descreve uma arquitetura de hardware para a implementação de uma versão sequencial do Filtro de Kalman Estendido (EKF, do inglês Extended Kalman Filter). Devido ao fato de que o EKF é computacionalmente intensivo, comumente ele é implementado em plataformas baseadas em PC (do inglês Personal Computer) para ser empregado em robótica móvel. Para permitir o desenvolvimento de plataformas robóticas pequenas (por exemplo, aquelas re- quisitadas em robótica móvel) condições especí cas tais como tamanho pequeno, consumo baixo de potência e capacidade de aritmética em ponto utuante são exigidos, assim como projetos de arquiteturas de hardware especí cas e adequadas. Desta maneira, a arquitetura proposta foi projetada para tarefas de auto-localização, usando operadores de aritmética de ponto utuante (em precisão simples), permitindo a fusão de dados provenientes de diferentes sensores tais como ultrassom e ladar. O sistema foi adaptado para ser aplicado em uma plataforma recon gurável, apropriada para tarefas de pesquisa, e a mesma foi testada em uma plataforma robótica Pioneer 3AT (da Mobile Robots Inc.) a m de avaliar sua funcionalidade, usando seu sistema de sen- soriamento. Para comparar o desempenho do sistema, o mesmo foi implementado em um PC, assim como pela utilização de um microprocessador embarcado na FPGA (o Nios II, da Altera). Neste trabalho, várias métricas foram utilizadas a m de avaliar o desempenho e a aplicabilidade do sistema, medindo o consumo de recursos na FPGA e seu desempenho. Devido ao fato de que este trabalho só está implementando a fase de atualização do EKF, o sistema geral foi testado assumindo que o robô está parado em uma posição previamente conhecida. ______________________________________________________________________________ ABSTRACT / This work describes a hardware architecture for implementing a sequential approach of the Extended Kalman Filter (EKF) that is suitable for mobile robotics tasks, such as self-localization, mapping, and navigation problems, especially when FPGAs (Field Programmable Gate Arrays) are used to execute this algorithm. Given that EKF is computationally intensive, commonly it is implemented in PC-based platforms to be employed on mobile robots. In order to allow the development of small robotic platforms (for instance those required in microrobotics area) speci c requirements such as small size, low-power, and oating-point arithmetic capability are demanded, as well as the design of speci c and suitable hardware architectures. Therefore, the proposed architecture has been achieved for self-localization task, using oating-point arithmetic operators (in simple precision), allowing the fusion of data coming from di erent sensors such as ultrasonic and laser range nder. The system has been adapted for achieving a recon gurable platform, suitable for research tasks, and the same has been tested in a Pioneer 3AT mobile robot platform (from Mobile Robots Inc.) for evaluating its functionality by using its local sensing system. In order to compare the performance of the system, the same localization technique has been implemented in a PC, as well as using an FPGA-embedded microprocessor (the Nios II from Altera Inc.) In this work several metrics have been used in order to evaluate the system performance and suitability, measuring both the FPGA resources consumption and performance. Given that in this work only the update phase of the EKF has been implemented the overall system has been tested assuming that the robot is stopped in a previously well-known position. ______________________________________________________________________________ RESUMEN / Este trabajo describe una arquitectura de hardware para la implementación de una versión secuencial del ltro de Kalman extendido (EKF del ingles Extended Kalman Filter). Debido al hecho de que el EKF es computacionalmente intensivo, típicamente es implementado en plataformas basadas en PC's (del ingles Personal Computer) para ser utilizado en robótica móvil. Para per- mitir el desarrollo de pequeñas plataformas robóticas(como las requeridas en robótica móvil) son exigidos condiciones especi cas como su pequeño tamaño, bajo consumo de potencia y capacidad de aritmética en punto otante, así como arquitecturas de hardware especi cas y adecuadas. De esta manera la arquitectura propuesta fue proyectada para tareas de auto-localización, usando operadores de aritmética de punto otante (en precisión simple), permitiendo la fusión de datos provenientes de diferentes sensores tales como ultrasonido y ladar. El sistema fue adaptado para aplicarlo en una plataforma recon gurable, apropiada para investigación, y la misma fue probada en una plataforma robótica denominada Pioneer 3AT (de la compañía Mobile Robots Inc.) utilizando el sistema de sensoramiento de este, con el propósito de validar su funcionalidad. Para comparar el desempeño del sistema, este fue implementado en un PC, así como en un microprocesador embarcado en una FPGA (Nios II, de Altera). En este trabajo, varias métricas fueron utilizadas con el propósito de validar el desempeño y la aplicabilidad del sistema, midiendo el consumo de recursos en la FPGA y su desempeño. Debido al hecho de que en el trabajo solo esta implementado la fase de actualizacion del EKF el sistema general fue probado asumiendo que el robot esta parado en una posición previamente conocida.

The introduction of digital filtering is a new and exciting approach in analyzing in-situ seismic data. Digital filters are also in the same spirit as the electric cone which replaced the mechanical cone in CPT* testing. That is, it is desirable to automate CPT testing in order to make it less operator dependent and increase the reliability and accuracy. In CPT seismic cone testing seismic waves are generated at the surface and recorded downhole with velocity or acceleration transducers. The seismic receivers record the different seismic wavelets (e.g., SV-waves, P-waves) allowing one to determine shear and compression wave velocities. In order to distinguish the different seismic events, an instrument with fast response time is desired (i.e., high natural frequency and low damping). This type of instrument is characteristic of an accelerometer. The fast response time (small time constant) of an accelerometer results in a very sensitive instrument with corresponding noisy time domain characteristics. One way to separate events is to characterize the signal frequencies and remove unwanted frequencies. Digital filtering is ideal for this application. The techniques of digital filtering introduced in this research are based on frequency domain filtering, where Fast Fourier, Butterworth Filter, and crosscorrelation algorithms are implemented. One based on time domain techniques, where a Kalman Filter is designed to model'the instrument and the physical environment. The crosscorrelation method allows one to focus on a specific wavelet and use all the information of the wavelets present averaging out any noises or irregularities and relying upon dominant responses. The Kalman Filter was applied in a manner in which it modelled the sensors used and the physical environment of the body waves and noise generation. The KF was investigated for its possible application to obtaining accurate estimates on the P-wave and S-wave amplitudes and arrival times. The KF is a very flexible tool which allows one to model the problem considered accurately. In addition, the KF works in the time domain which removes many of the limitations of the frequency domain techniques. The crosscorrelation filter concepts are applied by a program referred to as CROSSCOR. CROSSCOR is a graphics interactive program which displays the frequency spectrums, unfiltered and filtered time series and crosscorrelations on a mainframe graphics terminal which has been adapted to run on the IBM P.C. CROSSCOR was tested for performance by analyzing synthetic and real data. The results from the analysis on both synthetic and real data indicate that CROSSCOR is an accurate and user friendly tool which greatly assists one in obtaining seismic velocities. The performance of the Kalman Filter was analyzed by generating a source wavelet and passing it through the second order instrumentation. The second order response is then fed into the KF with the arrival time and maximum amplitude being determined. The filter was found to perform well and it has much promise in respect that if it is finely turned, it would be possible to obtain arrival times and amplitudes on line resulting in velocities and damping characteristics, respectively. * Cone Penetration Test / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

Orientador : Manuel de Jesus Mendes / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia de Campinas / Made available in DSpace on 2018-07-17T02:00:38Z (GMT). No. of bitstreams: 1 Bruno_PaulodeTarsoMartins_M.pdf: 1851397 bytes, checksum: e4163e65201b889c962b8fe6046de839 (MD5) Previous issue date: 1976 / Resumo: O problema da estimação do estado de um sistema dinâmico estocásticos, a partir de observações na saída, é de grande importância em engenharia. A partir de 1960 grandes impulso tem sido dado na solução das mais diferentes situações, encontrando-se atualmente grande serie de algoritmos de filtragem seqüencial. No presente trabalho estudam-se os quatro filtros de segunda ordem citados na literatura, analisando suas vantagens e desvantagens na aplicação a um sistema escalar; um desses algoritmos é aplicado a um problema pra tico e os resultados são comentados / Abstract: Not informed / Mestrado / Mestre em Engenharia Elétrica

This study is based on applying a non-linear mapping method, here the unscented Kalman filter; to estimate and optimize data rate resulting from the arrival rate having a Poisson distribution in an orthogonal frequency division multiplexing (OFDM) transmission system. OFDM is an emerging multi-carrier modulation scheme. With the growing need for quality of service in wireless communications, it is highly necessary to optimize resources in such a way that the overall performance of the system models should rise while keeping in mind the objective to achieve high data rate and efficient spectral methods in the near future. In this study, the results from the OFDM-TDMA transmission system have been used to apply cross-layer optimization between layers so as to treat different resources between layers simultaneously. The main controller manages the transmission of data between layers using the multicarrier modulation techniques. The unscented Kalman filter is used here to perform nonlinear mapping by estimating and optimizing the data rate, which result from the arrival rate having a Poisson distribution.

Estimation methods such as the Kalman filter identify best state estimates based on certain optimality criteria using a model of the system and the observations. A common assumption underlying the estimation is that the noise is Gaussian. In practical systems though, one quite frequently encounters thick-tailed, non-Gaussian noise. Statistically, contamination by this type of noise can be seen as inducing outliers among the data and leads to significant degradation in the KF. While many nonlinear methods to cope with non-Gaussian noise exist, a filter that is robust in the presence of outliers and maintains high statistical efficiency is desired. To solve this problem, a new robust Kalman filter framework is proposed that bounds the influence of observation, innovation, and structural outliers in a discrete linear system. This filter is designed to process the observations and predictions together, making it very effective in suppressing multiple outliers. In addition, it consists of a new prewhitening method that incorporates a robust multivariate estimator of location and covariance. Furthermore, the filter provides state estimates that are robust to outliers while maintaining a high statistical efficiency at the Gaussian distribution by applying a generalized maximum likelihood-type (GM) estimator. Finally, the filter incorporates the correct error covariance matrix that is derived using the GM-estimator's influence function. This dissertation also addresses robust state estimation for systems that follow a broad class of nonlinear models that possess two or more equilibrium points. Tracking state transitions from one equilibrium point to another rapidly and accurately in such models can be a difficult task, and a computationally simple solution is desirable. To that effect, a new robust extended Kalman filter is developed that exploits observational redundancy and the nonlinear weights of the GM-estimator to track the state transitions rapidly and accurately. Through simulations, the performances of the new filters are analyzed in terms of robustness to multiple outliers and estimation capabilities for the following applications: tracking autonomous systems, enhancing actual speech from cellular phones, and tracking climate transitions. Furthermore, the filters are compared with the state-of-the-art, i.e. the <i>H_â</i>-filter for tracking an autonomous vehicle and the extended Kalman filter for sensing climate transitions. / Ph. D.