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.

The conventional method of thermal modeling of space payloads is expensive and cumbersome. Radiation plays an important part in the thermal modeling of space payloads because of the presence of vacuum and deep space viewing. This induces strong nonlinearities into the thermal modeling process. There is a need for extensive correlation between the model and test data. This thesis presents Online Parameter Estimation as an approach to automate the thermal modeling process. The extended Kalman fillter (EKF) is the most widely used parameter estimation algorithm for nonlinear models. The unscented Kalman filter (UKF) is a new and more accurate technique for parameter estimation. These parameter estimation techniques have been evaluated with respect to data from ground tests conducted on an experimental space payload.

Made available in DSpace on 2014-06-11T19:26:55Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-06-29Bitstream added on 2014-06-13T20:27:33Z : No. of bitstreams: 1 bedutti_ad_me_sjrp.pdf: 1999751 bytes, checksum: 437383410c3f4cc28c116c28e9f7054a (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / O controle de plantas daninhas constitui um dos principais desafios no cultivo de área agrícolas. Quando presentes em quantidades descontroladas, estas plantas geram a diminuição na produtividade e ocasionam perdas significativas e indesejáveis. As perdas, aliadas ao alto custo de controle, motivam o desenvolvimento de ferramentas no auxílio a tomada de decisão, como mapas da distribuição de daninhas, visando o manejo localizado de herbicidas. Neste trabalho, considera-se a aplicação de um modelo espaço-temporal para a construção de mapas da distribuição de sementes de plantas daninhas em uma área agrícola de plantação de milho (Zea mays). Foram analisados dados reais, para as espécies Digitaria ciliaris, Euphorbia heterophilla L., Cenchrus echinatus L. e Bidens Pilosa L. e tamb´em dados simulados. O modelo envolve a combinação de estimação por krigagem e o filtro de Kalman. / The control of weeds is a major challenge in cultivation of agricultural areas. When present in uncontrolled quantities, these plants generate a decrease in productivity and cause significant and undesirable losses. The losses, combined with the high cost of control, motivate the development of tools to aid in taking decision, as maps of distribution of weed, to located handling of herbicides. In this work, was considered the application of a spatial-temporal model for construction of distribution maps of seed weeds in an agricultural area of corn plantation (Zea mays). Were analyzed real data, for the species Digitaria ciliaris, Euphorbia heterophilla L., Cenchrus echinatus L. and Bidens Pilosa L., and also simulated data. The model involves a combination of kriging estimation and Kalman filter.

Tsunami concerns have increased in the world after the 2004 Indian Ocean tsunami and the 2011 Tohoku tsunami. Consequently, tsunami models have been developed rapidly in the last few years. One of the advanced tsunami models is the GeoClaw tsunami model introduced by LeVeque (2011). This model is adaptive and consistent. Because of different sources of uncertainties in the model, observations are needed to improve model prediction through a data assimilation framework. Model inputs are earthquake parameters and topography. This thesis introduces a real-time tsunami forecasting method that combines tsunami model with observations using a hybrid ensemble Kalman filter and ensemble Kalman smoother. The filter is used for state prediction while the smoother operates smoothing to estimate the earthquake parameters. This method reduces the error produced by uncertain inputs. In addition, state-parameter EnKF is implemented to estimate earthquake parameters. Although number of observations is small, estimated parameters generates a better tsunami prediction than the model. Methods and results of prediction experiments in the Red Sea are presented and the prospect of developing an operational tsunami prediction system in the Red Sea is discussed.

When operating an aircraft, inaccurate measurements can have devastating consequences. For example, when measuring airspeed using a pitot tube, icing effects and other faults can result in erroneous measurements. Therefore, this master thesis aims to create an alternative method which utilizes known flight mechanical equations and sensor fusion to create an estimate of the airspeed during flight. For validation and generation of flight data, a simulation model developed by SAAB AB, called ARES, is used.  Two models are used to describe the aircraft behavior. One of which is called the dynamic model and utilizes forces acting upon the aircraft body in the equations of motion. The other model, called the kinematic model, instead describes the motion with accelerations of the aircraft body. The measurements used are the angle of attack (AoA), side-slip angle (SSA), GPS velocities, and angular rates from an inertial measurement unit (IMU). The dynamic model assumes that engine thrust and aerodynamic coefficients are already estimated to calculate resulting forces, meanwhile the kinematic model instead uses body fixed accelerations from the IMU. These models are combined with filters to create estimations of the airspeed. The filters used are the extended Kalman filter (EKF) and unscented Kalman filter (UKF). These are combined with the two models to create in total four methods to estimate the airspeed.  The results show no major difference in the performance between the filters except for computational time, for which the EKF has the fastest. Further, the result show similar airspeed estimation performance between the models, but differences can be seen. The kinematic model manages to estimate the wind with higher details and to converge faster, compared to the dynamic model. Both models suffer from an observability problem. This problem entails that the aircraft needs to be maneuvered to excite the AoA and SSA in order for the estimation methods to evaluate the wind, which is crucial for accurate airspeed estimation. The robustness of the dynamic model regarding errors in engine thrust and aerodynamic coefficients are also researched, which shows that the model is quite robust against errors in these values.

ITC/USA 2006 Conference Proceedings / The Forty-Second Annual International Telemetering Conference and Technical Exhibition / October 23-26, 2006 / Town and Country Resort & Convention Center, San Diego, California / OFDM is a preferred physical layer for an increasing number of telemetry and LAN applications. However, joint estimation of the multipath channel and frequency offset in OFDM remains a challenging problem. The Unscented Kalman Filter (UKF) is presented to solve the offset/channel tracking problem. The advantages of the UKF are that it is less susceptible to divergence than the EKF, and does not require computation of a Jacobian matrix. A hybrid analysis/simulation approach is developed to rapidly evaluate UKF performance in terms of symbol-error rate and channel/offset error for the 802.11a OFDM format.

No description available.

No description available.

Tesis para optar al grado de Magíster en Finanzas / Utilizando valores de cierres semanales, correspondientes al período comprendido entre el 28 de Agosto de 2000 al 14 de Agosto de 2006, se analiza la eficiencia de modelos multivariables dinámicos, optimizados por algoritmos genéticos y filtro de kalman, para predecir el signo de las variaciones semanales en la cotización bursátil de GE, GM, IBM, UTX, VZ, DJI e IPC. Los resultados fueron comparados con los de un modelo AR(1) y de un modelo multivariable ARIMAX(2,2,2). Los mejores modelos producidos por el algoritmo genético arrojaron un porcentaje de predicción de signo (PPS), para un conjunto extramuestral de 52 datos semanales, de un 77%, 71%, 81%, 75%, 75%, 81% y 77%, para las acciones GE, GM, IBM, UTX, VZ, DJI e IPC, respectivamente. La capacidad predictiva resultó significativa en cada una de las acciones, de acuerdo al test de acierto direccional de Pesaran & Timmerman (1992). Al analizar el PPS de los modelos de filtro de kalman, se encontró que estos fueron menores, resultando significativos en el caso de GE, GM, IBM, UTX y DJI. Por otro lado, el PPS de los modelos AR(1), se encontró que estos fueron no significativos para todas las acciones en estudio. Los modelos multivariables ARIMAX(2,2,2) registraron un PPS más alto que, los de filtro de kalman para el caso de UTX e IPC, siendo el primero no significativo. Además, los modelos construidos por el algoritmo genético generaron en promedio el mayor retorno acumulado corregido por riesgo, medido por los índices de Sharpe y Treynor, a excepción de GM e IPC, donde la rentabilidad más alta fue registrada por el modelo de filtro de kalman. Los resultados se confirman en las series generadas a través de un proceso bootstrap. De esta manera, se presenta evidencia de que, para el caso norteamericano, los modelos de algoritmos genéticos pueden predecir el cambio direccional del precio, junto con generar mayores retornos que un modelo ingenuo y una estrategia buy & hold. Lo anterior apoya las conclusiones del estudio de Leung, Daouk y Chen (2000), según el cual la predicción de la dirección del movimiento puede arrojar mayores ganancias de capital que la proyección del valor de cierre.

Today in most vehicles in battle, a camera system is used to manually lock a target and maintaining visual of the target as the vehicle is moving. In order to simplify this, this thesis investigates the approach to semi-automate the process by first manually locking the target and then let the camera approximate the trajectory of the enemy vehicle. In this thesis, the enemy vehicle is not moving. The ability to provide a truthful simulation environment for testing is crucial and will be discussed in this thesis along with three different estimators derived from the Kalman filter. Parameter identification and dynamic modelling of the camera are also presented that serves as a basis for the part of automatic control and for the experiments on the hardware. The simulation environment gave promising results when locating the target based on angle and radius estimation. By simulating a human operator, big deviations from the true value was no longer a problem since its purpose is to take over and steer the camera to the correct value. After gathering results from the simulations, Model-Based Design made it possible to test the algorithms in real life. The biggest challenge was to produce lifelike motions to test the hardware on and therefore made it harder to conclude the end result for the experiments carried out by the hardware on the moving platform.