Sequential decision making in general state space models

Many problems in control and signal processing can be formulated as sequential decision problems for general state space models. However, except for some simple models one cannot obtain analytical solutions and has to resort to approximation. In this thesis, we have investigated problems where Sequential Monte Carlo (SMC) methods can be combined with a gradient based search to provide solutions to online optimisation problems. We summarise the main contributions of the thesis as follows. Chapter 4 focuses on solving the sensor scheduling problem when cast as a controlled Hidden Markov Model. We consider the case in which the state, observation and action spaces are continuous. This general case is important as it is the natural framework for many applications. In sensor scheduling, our aim is to minimise the variance of the estimation error of the hidden state with respect to the action sequence. We present a novel SMC method that uses a stochastic gradient algorithm to find optimal actions. This is in contrast to existing works in the literature that only solve approximations to the original problem. In Chapter 5 we presented how an SMC can be used to solve a risk sensitive control problem. We adopt the use of the Feynman-Kac representation of a controlled Markov chain flow and exploit the properties of the logarithmic Lyapunov exponent, which lead to a policy gradient solution for the parameterised problem. The resulting SMC algorithm follows a similar structure with the Recursive Maximum Likelihood(RML) algorithm for online parameter estimation. In Chapters 6, 7 and 8, dynamic Graphical models were combined with with state space models for the purpose of online decentralised inference. We have concentrated more on the distributed parameter estimation problem using two Maximum Likelihood techniques, namely Recursive Maximum Likelihood (RML) and Expectation Maximization (EM). The resulting algorithms can be interpreted as an extension of the Belief Propagation (BP) algorithm to compute likelihood gradients. In order to design an SMC algorithm, in Chapter 8 uses a nonparametric approximations for Belief Propagation. The algorithms were successfully applied to solve the sensor localisation problem for sensor networks of small and medium size.

On the computation of space-time correlations by large-eddy simulation

The effect of subgrid-scale (SGS) modeling on velocity (space-) time correlations is investigated in decaying isotropic turbulence. The performance of several SGS models is evaluated, which shows superiority of the dynamic Smagorinsky model used in conjunction with the multiscale large-eddy simulation (LES) procedure. Compared to the results of direct numerical simulation, LES is shown to underpredict the (un-normalized) correlation magnitude and slightly overpredict the decorrelation time scales. This can lead to inaccurate solutions in applications such as aeroacoustics. The underprediction of correlation functions is particularly severe for higher wavenumber modes which are swept by the most energetic modes. The classic sweeping hypothesis for stationary turbulence is generalized for decaying turbulence and used to analyze the observed discrepancies. Based on this analysis, the time correlations are determined by the wavenumber energy spectra and the sweeping velocity, which is the square root of the total energy. Hence, an accurate prediction of the instantaneous energy spectra is most critical to the accurate computation of time correlations. (C) 2004 American Institute of Physics.

An overview of sequential Monte Carlo methods for parameter estimation in general state-space models

Nonlinear non-Gaussian state-space models arise in numerous applications in control and signal processing. Sequential Monte Carlo (SMC) methods, also known as Particle Filters, are numerical techniques based on Importance Sampling for solving the optimal state estimation problem. The task of calibrating the state-space model is an important problem frequently faced by practitioners and the observed data may be used to estimate the parameters of the model. The aim of this paper is to present a comprehensive overview of SMC methods that have been proposed for this task accompanied with a discussion of their advantages and limitations.

Finding space to mental health : promoting mental health in adolescents (12 to 14 year olds): preliminary results of the intervention's effectiveness

Resumen: La salud mental y el bienestar son fundamentales para nuestra capacidad colectiva y individual como seres humanos de pensar, de exteriorizar los sentimientos, de establecer y mantener relaciones, para estudiar, para perseguir las actividades de ocio, para tomar decisiones diarias y para disfrutar de una vida plena. Una adolescencia saludable es un prerrequisito para una vida adulta saludable. Sin embargo, la realidad actual presenta un panorama preocupante. La formación del capital mental individual y colectivo - especialmente en las primeras etapas de la vida - está siendo retenida por una serie de riesgos evitables para la salud mental (World Health Organization [WHO], 2013). Los adolescentes del sur de Europa (región que ha sido más severamente afectada por la crisis financiera; e.g., Portugal) son señalados como un grupo extremadamente vulnerable, ya que su salud mental fácilmente podría ser influenciada por las dificultades económicas de sus padres y la escasez de solidaridad social (European Parliament, 2012). La promoción de la salud mental de los adolescentes es considerada como una preocupación fundamental (WHO, 2005a, 2013). En este ámbito, las intervenciones centradas en la promoción de la literacía de la salud mental han revelado importantes ventajas en la prevención, reconocimiento, intervención precoz y la reducción del estigma (Pinfold, Stuart, Thornicroft & Arboleda-Florez, 2005; Pinfold, Toulmin, Thornicroft, Huxley, Farmer & Graham, 2003; Schulze, Richter-Werling, Matschinger & Angermeyer, 2003; Stuart, 2006). En consonancia con los marcos de promoción de la salud mentales propuestos por la Organización Mundial de la Salud (2005a), tenemos que involucrar a jóvenes en los ambientes donde interactúan (Burns, 2011). Las escuelas son implícitamente uno de los locales más importantes para la promoción de la salud mental de los adolescentes (Barry, Clarke, Jenkins & Patel, 2013; WHO, 2001). El proyecto “Abrir Espacio para la Salud Mental – Promoción de la salud mental en adolescentes (12-14 años)” tiene como objetivo incrementar literacía de la salud mental en los jóvenes. En el primer año se ha desarrollado un instrumento de evaluación - Mental Health Literacy questionnaire (MHLq) - y la intervención para la promoción de la salud mental. La intervención consiste en 2 sesiones, 90 minutos cada una, implementadas con intervalo de una semana. Siguen una metodología interactiva, utilizando dinámicas de grupo, videos, música y discusión. El estudio de la eficacia de la intervención se lleva a cabo mediante un análisis pre y pos-test con el MHLq, utilizando un grupo experimental y un grupo de control. Este artículo presenta los resultados preliminares de la eficacia de la intervención de promoción de la salud mental en una muestra de 100 adolescentes portugueses (12-14 años). El pos-test mostró un incremento de los niveles de conocimientos de salud mental y estrategias de autoayuda. Los resultados sugieren que la intervención desarrollada parece ser adecuada al objetivo propuesto y refuerzan la creencia de que intervenciones escolares, sistemáticas y sostenibles, para la promoción de la salud mental con jóvenes, es un enfoque prometedor para la promoción de la literacía de la salud mental (Schulze et al., 2003; Rickwood et al., 2005; Corrigan et al., 2007; WHO, 2010).

Characteristics of deep penetration laser welding of dissimilar metal Ni-based cast superalloy K418 and alloy steel 42CrMo

Experimental trials of autogenous deep penetration welding between dissimilar cast Ni-based superalloy K418 and alloy steel 42CrMo flat plates with 5.0 mm thickness were conducted using a 3 kW continuous wave (CW) Nd:YAG laser. The influences of laser output power, welding velocity and defocusing distance on the morphology, welding depth and width as well as quality of the welded seam were investigated. Results show that full keyhole welding is not formed on both K4.18 and 42CrMo side, simultaneously, due to the relatively low output power. Partial fusion is observed on the welded seam near 42CrMo side because of the large disparity of thermal-physical and high-temperature mechanical properties of these two materials. Tile rnicrohardness of the laser-welded joint was also examined and analyzed. It is suggested that applying negative defocusing in the range of Raylei length can increase the welding depth and improve tile coupling efficiency of the laser materials interaction. (c) 2007 Elsevier Ltd. All rights reserved.

Stress intensity factors calculation in anti-plane fracture problem by orthogonal integral extraction method based on femol

For an anti-plane problem, the differential operator is self-adjoint and the corresponding eigenfunctions belong to the Hilbert space. The orthogonal property between eigenfunctions (or between the derivatives of eigenfunctions) of anti-plane problem is exploited. We developed for the first time two sets of radius-independent orthogonal integrals for extraction of stress intensity factors (SIFs), so any order SIF can be extracted based on a certain known solution of displacement (an analytic result or a numerical result). Many numerical examples based on the finite element method of lines (FEMOL) show that the present method is very powerful and efficient.

Is there chaotic synchronization in space extend systems

Based on coupled map lattice (CML), the chaotic synchronous pattern in space extend systems is discussed. Making use of the criterion for the existence and the conditions of stability, we find an important difference between chaotic and nonchaotic movements in synchronization. A few numerical results are presented.

One-dimensional improvement and two-dimensional and 3-dimensional treatments for stable oscillation condition, mode structure and power output in high-speed-flow lasers

For high-speed-flow lasers, the one-dimensional and first-order approximate treatment in[1] under approximation of geometrical optics is improved still within the scope of approx-imation of geometrical optics. The strict accurate results are obtained, and what is more,two- and three-dimensional treatments are done. Thus for two- and three-dimensional cases, thestable oscillation condition, the formulae of power output and analytical expression of modesunder approximation of geometrical optics (in terms of gain function) are derived. Accord-ing to the present theory, one-and two-dimensional calculations for the typical case of Gerry'sexperiment are presented. All the results coincide well with the experiment and are better thanthe results obtained in [1].In addition, the applicable scope of Lee's stable oscillation condition given by [1] is ex-panded; the condition for the approximation of gcometrical optics to be applied to mode con-structure in optical cavity is obtained for the first time and the difference between thiscondition and that for free space is also pointed out in the present work.