L'apnée est une complication fréquente chez les nouveaux-nés prématurés. L'un des problèmes les plus fréquents est l'épisode d'apnée bradycardie dont la répétition influence de manière négative le développement de l'enfant. C'est pourquoi les enfants prématurés sont surveillés en continu par un système de monitoring. Depuis la mise en place de ce système, l'espérance de vie et le pronostic de vie des prématurés ont été considérablement améliorés et ainsi la mortalité réduite. En effet, les avancées technologiques en électronique, informatique et télécommunications ont conduit à l'élaboration de systèmes multivoies de monitoring néonatal de plus en plus performants. L'un des principaux signaux exploités dans ces systèmes est l'électrocardiogramme (ECG). Toutefois, même si l'analyse de l'ECG a évolué au fil des années, l'ensemble des informations qu'il fournit n'est pas encore totalement exploité dans les processus de décision, notamment en monitoring en Unité de Soins Intensifs en Néonatalogie (USIN). L'objectif principal de cette thèse est d'améliorer la prise en compte des dynamiques multi-dimensionnelles en proposant de nouvelles approches basées sur un formalisme bayésien, pour la détection précoce des apnées bradycardies chez le nouveau-né prématuré. Aussi, dans cette thèse, nous proposons deux approches bayésiennes, basées sur les caractéristiques de signaux biologiques en vue de la détection précoce de l'apnée bradycardie des nouveaux-nés prématurés. Tout d'abord avec l'approche de Markov caché, nous proposons deux extensions du Modèle de Markov Caché (MMC) classique. La première, qui s'appelle Modèle de Markov Caché Couplé (MMCC), créé une chaîne de Markov à chaque dimension de l'observation et établit un couplage entre les chaînes. La seconde, qui s'appelle Modèle Semi-Markov Caché Couplé (MSMCC), combine les caractéristiques du modèle de MSMC avec le mécanisme de couplage entre canaux. Pour les deux nouveaux modèles (MMCC et MSMCC), les algorithmes récursifs basées sur la version classique de Forward-Backward sont introduits pour résoudre les problèmes d'apprentissage et d'inférence dans le cas couplé. En plus des modèles de Markov, nous proposons deux approches passées sur les filtres de Kalman pour la détection d'apnée. La première utilise les modifications de la morphologie du complexe QRS et est inspirée du modèle générateur de McSharry, déjà utilisé en couplant avec un filtre de Kalman étendu dans le but de détecter des changements subtils de l'ECG, échantillon par échantillon. La deuxième utilise deux modèles AR (l'un pour le processus normal et l'autre pour le processus de bradycardie). Les modèles AR sont appliqués sur la série RR, alors que le filtre de Kalman suit l'évolution des paramètres du modèle AR et fournit une mesure de probabilité des deux processus concurrents. / Apnea-bradycardia episodes (breathing pauses associated with a significant fall in heart rate) are the most common disease in preterm infants. Consequences associated with apnea-bradycardia episodes involve a compromise in oxygenation and tissue perfusion, a poor neuromotor prognosis at childhood and a predisposing factor to sudden-death syndrome in preterm newborns. It is therefore important that these episodes are recognized (early detected or predicted if possible), to start an appropriate treatment and to prevent the associated risks. In this thesis, we propose two Bayesian Network (BN) approaches (Markovian and Switching Kalman Filter) for the early detection of apnea bradycardia events on preterm infants, using different features extracted from electrocardiographic (ECG) recordings. Concerning the Markovian approach, we propose new frameworks for two generalizations of the classical Hidden Markov Model (HMM). The first framework, Coupled Hidden Markov Model (CHMM), is accomplished by assigning a Markov chain (channel) to each dimension of observation and establishing a coupling among channels. The second framework, Coupled Hidden semi Markov Model (CHMM), combines the characteristics of Hidden semi Markov Model (HSMM) with the above-mentioned coupling concept. For each framework, we present appropriate recursions in order to use modified Forward-Backward (FB) algorithms to solve the learning and inference problems. The proposed learning algorithm is based on Maximum Likelihood (ML) criteria. Moreover, we propose two new switching Kalman Filter (SKF) based algorithms, called wave-based and R-based, to present an index for bradycardia detection from ECG. The wave-based algorithm is established based on McSarry's dynamical model for ECG beat generation which is used in an Extended Kalman filter algorithm in order to detect subtle changes in ECG sample by sample. We also propose a new SKF algorithm to model normal beats and those with bradycardia by two different AR processes.

The Kalman Filter has been applied to many fields of hydrology, particularly in the area of flood forecasting. This recursive estimation technique is based on a state-space approach which combines model description of a process with data information, and accounts for uncertainties in a hydrologic system. This thesis deals with applications of the Kalman Filter to ARMAX models in the context of streamflow prediction. Implementation of the Kalman Filter requires specification of the noise covariances (Q, R) and initial conditions of the state vector (x₀, P₀). Difficulties arise in streamflow applications because these quantities are often not known. Forecasting performance of the Kalman Filter is examined using synthetic flow data, generated with chosen values for the initial state vector and the noise covariances. An ARMAX model is cast into state-space form with the coefficients as the state vector. Sensitivity of the flow forecasts to specification of x₀, P₀, Q, R, (which may be different from the generation values) is examined. The filter's forecasting performance is mainly affected by the combined specification of Q and R. When both noise covariances are unknown, they should be specified relatively large in order to achieve a reasonable forecasting performance. Specififying Q too small and R too large should be avoided as it results in poor flow forecasts. The filter's performance is also examined using actual flow data from a large river, whose behavior changes slowly with time. Three simple ARMAX models are used for this investigation. Although there are different ways of writing the ARMAX model in state-space form, it is found that the best forecasting scheme is to model the ARMAX coefficients as the state vector. Under this formulation, the Kalman Filter is used to give recursive estimates of the coefficients. Hence flow predictions can be revised at each time step with the latest state estimate. This formulation also has the feature that initial values of the ARMAX coefficients need not be known accurately. The noise variances of each of the three models are estimated by the method of maximum likelihood, whereby the likelihood function is evaluated in terms of the innovations. Analyses of flow data for the stations considered in this thesis, indicate that the variance of the measurement error is proportional to the square of the flow. In practice, flow predictions several time steps in advance are often required. For autoregressive processes, this involves unknown elements in the system matrix H of the Kalman model. The Kalman algorithm underestimates the variance of the forecast error if H and x are both unknown. For the AR(1) model, a general expression for the mean square error of the forecast is developed. It is shown that the formula reduces to the Kalman equation for the case where the system matrix is known. The importance of this formula is realized in forecasting situations where management decisions depend on the reliability of flow predictions, reflected by their mean square errors. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

Magíster en Ciencias de la Ingeniería, Mención Eléctrica / Ingeniero Civil Electricista / Esta tesis se centra en el estudio del estado de carga (SoC: state of charge) para acumuladores de energía del tipo plomo ácido. Este parámetro es de suma importancia en aplicaciones donde el acumulador está sometido permanentemente a situaciones de carga y/o descarga, como las que se presentan en vehículos eléctricos o micro-redes, por ejemplo. El SoC se define como la energía expresada como un porcentaje de la capacidad nominal, que aún está disponible para ser utilizada. Este indicador depende de muchas otras magnitudes, tales como: temperatura, tasas de corriente de carga/descarga, tiempo de uso, histéresis, y auto descarga. Este parámetro no es medible, por lo cual es necesario estimarlo en base a mediciones de otras señales disponibles en los acumuladores, tales como tensión, corriente y temperatura. Para desarrollar un estimador del SoC hay que tener en cuenta dos aspectos; el primero de ellos es contar con un buen modelo que represente el comportamiento del acumulador de manera adecuada, mientras que el segundo, dice relación con el algoritmo utilizado para realizar la estimación. Ambos aspectos suponen contar con información del acumulador para poder identificar el modelo y diseñar el estimador. Por lo tanto, se establecieron los objetivos de la tesis, los cuales corresponden en primer término a la construcción de un prototipo con el cual se pueda someter el banco de acumuladores a diversos perfiles de carga/descarga. Luego en base a la información generada mediante este sistema experimental, derivar un modelo de baterías que sea sencillo de implementar y requiera poca cantidad de información. Dicho modelo corresponde a un modelo difuso. Con el modelo de baterías ya definido, se utiliza el algoritmo del filtro extendido de Kalman para desarrollar un estimador del SoC basado en el modelo propuesto. Es importante destacar que tanto el modelo como el estimador son evaluados y comparados con modelos de baterías convencionales y con estimadores basados en ellos. (Implementados con el algoritmo de Kalman). Los aportes del trabajo de tesis, son en primer lugar, la construcción del sistema experimental, el cual servirá para otras investigaciones relacionadas a acumuladores de energía. En segundo término, se tiene que la metodología basada en lógica difusa (para el desarrollo del modelo), es novedosa, pues hasta el momento sólo ha sido implementada con datos basados en mediciones en el dominio de la frecuencia o en conjunto con redes neuronales. Lo que supone en el primer caso que el modelo no pueda ser llevado a la práctica debido al costoso equipamiento necesario para obtener datos en el dominio de la frecuencia, y en la necesidad de contar con gran cantidad de información para el segundo caso. Finalmente es importante mencionar que las baterías consideradas para esta tesis, están presentes en la micro-red Huatacondo perteneciente al Centro de Energía de la Universidad de Chile.

Orientadores: Neucimar Jeronimo Leite, Ricardo M. L Barros / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-07-23T22:55:06Z (GMT). No. of bitstreams: 1 FigueroaRivero_PascualJovino_M.pdf: 10223236 bytes, checksum: c66593c6d0899f3739e5a67519c38f63 (MD5) Previous issue date: 1998 / Resumo: Esta dissertação aborda o problema de acompanhamento (tracking) do movimento humano numa seqüência de imagens. O interesse pelo acompanhamento do movimento humano está relacionado com a análise e reconhecimento do movimento utilizado para diferentes aplicações, tais como, análise clínica, análise do movimento no esporte, ergonomia, coreografia de dança, etc. Em particular, nós abordamos este problema utilizando marcadores (adesivos) atados ou colados ao corpo humano, diferentemente de outros métodos que utilizam modelos do corpo humano e seu movimento. A utilização de marcadores permite obter resultados de maior precisão e que são de grande importância nas aplicações relacionadas com a biomecânica. Neste trabalho apresentamos algoritmos para a perseguição automática dos marcadores numa seqüência de imagens. O conjunto nestes algoritmos inclui a extração dos marcadores, a predição da posição nos próximos quadros, e sua identificação e classificação. O método discutido aqui permite a implementação de sistemas de análise do movimento humano baseados em câmeras normais, ao contrário dos sistemas que utilizam câmeras especiais, em que os custos são elevados. Um ponto importante deste trabalho é a introdução de ferramentas de processamento de imagens, em particular a morfologia matemática, para a extração dos marcadores. / Abstract: In this work we consider the problem of tracking human movements. The interest for this problem concerns the analysis and recognition of human motion used in a variety of applications, such as clinical gait analysis, improvement of the human performance in sports, ergonomics, choreography of dance, and so on. In particular, we treat this problem using markers attached or fixed to human body, different from the methods based on human body and their motion models. The use of markers allows us to get more precise results in applications, such as Biomechanics where this precision is very important. We present some algorithms for tracking markers in a sequence of images, consisting in: extraction (segmentation) of the markers, prediction of their position in the next frames, and their recognition and classification. The definition of these algorithms yields the implementation of systemts for human motion analysis based on normal CCD cameras, unlike the systems based on special optoelectronic devices. / Mestrado / Mestre em Ciência da Computação

Orientador : Clesio Luis Tozzi / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica / Made available in DSpace on 2018-07-18T15:25:33Z (GMT). No. of bitstreams: 1 Tommaselli_AntonioMariaGarcia_D.pdf: 7554867 bytes, checksum: 581a35b8d83d01a09e8bd0065ff6cb0b (MD5) Previous issue date: 1993 / Resumo: Este trabalho procura apresentar contribuições à resolução dos problemas de localização da câmara e de objetos, utilizando como apoio linhas retas conhecidas no espaço objeto e cujas homólogas tenham sido extraídas na imagem. Um modelo matemático do tipo explícito, e que permita a determinação simultânea dos seis parâmetros de orientação, é desenvolvido, sendo, então, tratado por Filtragem Kalman. A natureza recursiva do Filtro, possibilita a definição de um processo iterativo, em que as estimativas sequenciais dos parâmetros realimentam a etapa de extração de feições. Isto leva a uma gradativa redução do espaço de busca das feições na imagem, e à conseqüente diminuição do custo computacional. Pressupõe-se a disponibilidade de estimativas preditas de boa qualidade, em consonância com o conceito de Visão de Verificação. Conclui-se que, dentro deste contexto, a introdução de métodos de estimação de parâmetros, além de emprestar confiabilidade ao processo, permite a redução do custo computacional. Os procedimentos propostos foram implementados e testados com dados simulados e com dados reais. Os resultados obtidos são discutidos e analisados e as conclusões resultantes são apresentadas / Abstract: The aim of this work is to present contributions to the solution of the camera location and object location problems using known straight lines in the object space and extracted in the image. An explicit and simultaneous mathematical model using straight lines is derived, aiming Kalman Filtering application. The recursive nature of the Filter can be used to define an iterative process in which the sequentiall estimates feedback the feature extraction step. This leads to a gradual reduction of the features search space and, decreases. thus, computation time also The availability of good quali ty predicted estimates is assumed, according to the concept of Verification Vision. Besides the increase in reliability, estimation methods could also reduce the computational costs. The proposed approach was tested wi th simulated and real data. The obtained results are discussed and conclusions are presented. / Doutorado / Automação / Doutor em Engenharia Elétrica

Orientador: Clarice Azevedo de Luna Freire / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Ciencia da Computação / Made available in DSpace on 2018-07-19T14:12:26Z (GMT). No. of bitstreams: 1 Guimaraes_PauloRicardoBittencourt_M.pdf: 2127854 bytes, checksum: ebe87872a5069d20597ca0bce32b21d2 (MD5) Previous issue date: 1994 / Resumo: Não informado. / Abstract: Not informed. / Mestrado / Mestre em Estatística

Memoria para optar al título de Ingeniera Civil en Computación / El presente trabajo describe el desarrollo de un software en Python destinado a la detección de fenómenos astronómicos transitorios como las supernovas que corresponden a eventos caracterizados por un incremento rápido en su luminosidad y un consecuente decremento lento. El programa se diseñó sobre la base de una rutina ya implementada la cual hace uso de estimaciones generadas por métodos del filtro de Kalman: en su versión clásica (o básica) o su versión de máxima correntropı́a. Debido a que esta rutina presenta complicaciones en la administración de archivos y manejo de parámetros (producido principalmente por hard-coding) se realizó un proceso de refactoring que implica además diseñar y generar una nueva familia de filtros de Kalman basados en el patrón de diseño Strategy. Sobre este código refactorizado se efectuaron pruebas de rendimiento obteniéndose ası́ una mejora en términos de tiempo pero no en la memoria principal utilizada. Por otro lado se realizaron pruebas de detección usando el conjunto de 93 supernovas detectadas por el sondeo de HiTS del año 2015, hallándose mejoras notables en la disminución de falsos positivos ası́ como también un leve aumento en el número de verdaderos positivos al emplear las versiones clásica y de máxima correntropı́a de los filtros refactorizados. Sin embargo no ocurrió lo mismo con el nuevo filtro unscented, que permite emplear funciones no lineales al momento de estimar. Para este filtro se usaron una función cuadrática y otra de exponente 1,5; evaluadas sobre el paso del tiempo desde el inicio de las observaciones (o épocas). Se recomienda continuar estudiando el nuevo filtro de Kalman de aproximación no lineal debido al acotado conjunto de parámetros y funciones utilizado durante la realización de este trabajo.

Understanding the geology and the hydrology of the subsurface is important to model the fluid flow and the behavior of the contaminant. It is essential to have an accurate knowledge of the movement of the contaminants in the porous media in order to track them and later extract them from the aquifer. A two-dimensional flow model is studied and then applied on a linear contaminant transport model in the same porous medium. Because of possible different sources of uncertainties, the deterministic model by itself cannot give exact estimations for the future contaminant state. Incorporating observations in the model can guide it to the true state. This is usually done using the Kalman filter (KF) when the system is linear and the extended Kalman filter (EKF) when the system is nonlinear. To overcome the high computational cost required by the KF, we use the singular evolutive Kalman filter (SEKF) and the singular evolutive extended Kalman filter (SEEKF) approximations of the KF operating with low-rank covariance matrices. The SEKF can be implemented on large dimensional contaminant problems while the usage of the KF is not possible. Experimental results show that with perfect and imperfect models, the low rank filters can provide as much accurate estimates as the full KF but at much less computational cost. Localization can help the filter analysis as long as there are enough neighborhood data to the point being analyzed. Estimating the permeabilities of the aquifer is successfully tackled using both the EKF and the SEEKF.

The goal of this paper is to suggest suitable method for angle measuring of mobile robot. There are being analyzed 3 different sensors – gyroscope, accelerometer and electronic compass in the prologue. Their advantages and disadvantages in the theoretical way are being explained in this part and also their opportunities of use in the practical way. In the following parts the work is focused on MEMS gyroscopes and their opportunities of use in the practical way with regard to achievable exactness and to the application for development of its exactness. The application of device together with main SW for microcontroller and the valuation of achievable exactness and determined facts are being described in the conclusion part.

Kamerabasierte Verfahren zur Steuerung autonomer mobiler Roboter wurden in den letzten Jahren immer populärer. In dieser Arbeit wird der Einsatz eines Stereokamerasystems und eines Mono-Kamera-SLAM Verfahrens hinsichtlich der Unterstützung der Navigation eines autonomen Luftschiffes untersucht. Mit Hilfe von Sensordaten aus IMU, GPS und Kamera wird eine Positionsschätzung über eine Sensorfusion mit Hilfe des Extended und des Unscented Kalman Filters durchgeführt.