Models and algorithms for tracking of maneuvering objects using variable rate particle filters

Standard algorithms in tracking and other state-space models assume identical and synchronous sampling rates for the state and measurement processes. However, real trajectories of objects are typically characterized by prolonged smooth sections, with sharp, but infrequent, changes. Thus, a more parsimonious representation of a target trajectory may be obtained by direct modeling of maneuver times in the state process, independently from the observation times. This is achieved by assuming the state arrival times to follow a random process, typically specified as Markovian, so that state points may be allocated along the trajectory according to the degree of variation observed. The resulting variable dimension state inference problem is solved by developing an efficient variable rate particle filtering algorithm to recursively update the posterior distribution of the state sequence as new data becomes available. The methodology is quite general and can be applied across many models where dynamic model uncertainty occurs on-line. Specific models are proposed for the dynamics of a moving object under internal forcing, expressed in terms of the intrinsic dynamics of the object. The performance of the algorithms with these dynamical models is demonstrated on several challenging maneuvering target tracking problems in clutter. © 2006 IEEE.

Dasher - A data entry interface using continuous gestures and language models

Existing devices for communicating information to computers are bulky, slow to use, or unreliable. Dasher is a new interface incorporating language modelling and driven by continuous two-dimensional gestures, e.g. a mouse, touchscreen, or eye-tracker. Tests have shown that this device can be used to enter text at a rate of up to 34 words per minute, compared with typical ten-finger keyboard typing of 40-60 words per minute. Although the interface is slower than a conventional keyboard, it is small and simple, and could be used on personal data assistants and by motion-impaired computer users.

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.

Comparison of the HSPF and HBV-INCA models: Crossing the Atlantic divide

Observation shows that the watershed-scale models in common use in the United States (US) differ from those used in the European Union (EU). The question arises whether the difference in model use is due to familiarity or necessity. Do conditions in each continent require the use of unique watershed-scale models, or are models sufficiently customizable that independent development of models that serve the same purpose (e.g., continuous/event- based, lumped/distributed, field-Awatershed-scale) is unnecessary? This paper explores this question through the application of two continuous, semi-distributed, watershed-scale models (HSPF and HBV-INCA) to a rural catchment in southern England. The Hydrological Simulation Program-Fortran (HSPF) model is in wide use in the United States. The Integrated Catchments (INCA) model has been used extensively in Europe, and particularly in England. The results of simulation from both models are presented herein. Both models performed adequately according to the criteria set for them. This suggests that there was not a necessity to have alternative, yet similar, models. This partially supports a general conclusion that resources should be devoted towards training in the use of existing models rather than development of new models that serve a similar purpose to existing models. A further comparison of water quality predictions from both models may alter this conclusion.

Structured discriminative models for noise robust continuous speech recognition

Recently there has been interest in structured discriminative models for speech recognition. In these models sentence posteriors are directly modelled, given a set of features extracted from the observation sequence, and hypothesised word sequence. In previous work these discriminative models have been combined with features derived from generative models for noise-robust speech recognition for continuous digits. This paper extends this work to medium to large vocabulary tasks. The form of the score-space extracted using the generative models, and parameter tying of the discriminative model, are both discussed. Update formulae for both conditional maximum likelihood and minimum Bayes' risk training are described. Experimental results are presented on small and medium to large vocabulary noise-corrupted speech recognition tasks: AURORA 2 and 4. © 2011 IEEE.

Displacement and mixing ventilation driven by opposing wind and buoyancy

The effect of an opposing wind on the stratification and flow produced by a buoyant plume rising from a heat source on the floor of a ventilated enclosure is investigated. Ventilation openings located at high level on the windward side of the enclosure and at low level on the leeward side allow a wind-driven flow from high to low level, opposite to the buoyancy-driven flow. One of two stable steady flow regimes is established depending on a dimensionless parameter F that characterizes the relative magnitudes of the wind-driven and buoyancy-driven velocities within the enclosure, and on the time history of the flow. A third, unstable steady flow solution is identified. For small opposing winds (small F) a steady, two-layer stratification and displacement ventilation is established. Exterior fluid enters through the lower leeward openings and buoyant interior fluid leaves through the upper windward openings. As the wind speed increases, the opposing wind may cause a reversal in the flow direction. In this case, cool exterior fluid enters through the high windward openings and mixes the interior fluid, which exits through the leeward openings. There are now two possibilities. If the rate of heat input by the source exceeds the rate of heat loss through the leeward openings, the temperature of the interior increases and this flow reversal is only maintained temporarily. The buoyancy force increases with time, the flow reverts to its original direction, and steady two-layer displacement ventilation is re-established and maintained. In this regime, the increase in wind speed increases the depth and temperature of the warm upper layer, and reduces the ventilation flow rate. If, on the other hand, the heat loss exceeds the heat input, the interior cools and the buoyancy-driven flow decreases. The reversed flow is maintained, the stratification is destroyed and mixing ventilation occurs. Further increases in wind speed increase the ventilation rate and decrease the interior temperature. The transitions between the two ventilation flow patterns exhibit hysteresis. The change from displacement ventilation to mixing ventilation occurs at a higher F than the transition from mixing to displacement. Further, we find that the transition from mixing to displacement ventilation occurs at a fixed value of F, whereas the transition from displacement to mixing flow is dependent on the details of the time history of the flow and the geometry of the openings, and is not determined solely by the value of F. Theoretical models that predic t the steady stratification profiles and flow rates for the displacement and mixing ventilation, and the transitions between them, are presented and compared with measurements from laboratory experiments. The transition between these ventilation patterns completely changes the internal environment, and we discuss some of the implications for the natural ventilation of buildings. © 2004 Cambridge University Press.

Linear gaussian computations for near-exact Bayesian Monte Carlo inference in skewed alpha-stable time series models

In this paper we study parameter estimation for time series with asymmetric α-stable innovations. The proposed methods use a Poisson sum series representation (PSSR) for the asymmetric α-stable noise to express the process in a conditionally Gaussian framework. That allows us to implement Bayesian parameter estimation using Markov chain Monte Carlo (MCMC) methods. We further enhance the series representation by introducing a novel approximation of the series residual terms in which we are able to characterise the mean and variance of the approximation. Simulations illustrate the proposed framework applied to linear time series, estimating the model parameter values and model order P for an autoregressive (AR(P)) model driven by asymmetric α-stable innovations. © 2012 IEEE.

Bayesian parameter identification for vibro-acoustic models with nonparametric uncertainty

Vibration and acoustic analysis at higher frequencies faces two challenges: computing the response without using an excessive number of degrees of freedom, and quantifying its uncertainty due to small spatial variations in geometry, material properties and boundary conditions. Efficient models make use of the observation that when the response of a decoupled vibro-acoustic subsystem is sufficiently sensitive to uncertainty in such spatial variations, the local statistics of its natural frequencies and mode shapes saturate to universal probability distributions. This holds irrespective of the causes that underly these spatial variations and thus leads to a nonparametric description of uncertainty. This work deals with the identification of uncertain parameters in such models by using experimental data. One of the difficulties is that both experimental errors and modeling errors, due to the nonparametric uncertainty that is inherent to the model type, are present. This is tackled by employing a Bayesian inference strategy. The prior probability distribution of the uncertain parameters is constructed using the maximum entropy principle. The likelihood function that is subsequently computed takes the experimental information, the experimental errors and the modeling errors into account. The posterior probability distribution, which is computed with the Markov Chain Monte Carlo method, provides a full uncertainty quantification of the identified parameters, and indicates how well their uncertainty is reduced, with respect to the prior information, by the experimental data. © 2013 Taylor & Francis Group, London.

Kernelized log linear models for continuous speech recognition

Large margin criteria and discriminative models are two effective improvements for HMM-based speech recognition. This paper proposed a large margin trained log linear model with kernels for CSR. To avoid explicitly computing in the high dimensional feature space and to achieve the nonlinear decision boundaries, a kernel based training and decoding framework is proposed in this work. To make the system robust to noise a kernel adaptation scheme is also presented. Previous work in this area is extended in two directions. First, most kernels for CSR focus on measuring the similarity between two observation sequences. The proposed joint kernels defined a similarity between two observation-label sequence pairs on the sentence level. Second, this paper addresses how to efficiently employ kernels in large margin training and decoding with lattices. To the best of our knowledge, this is the first attempt at using large margin kernel-based log linear models for CSR. The model is evaluated on a noise corrupted continuous digit task: AURORA 2.0. © 2013 IEEE.