Model based learning of sigma points in unscented Kalman filtering

The unscented Kalman filter (UKF) is a widely used method in control and time series applications. The UKF suffers from arbitrary parameters necessary for a step known as sigma point placement, causing it to perform poorly in nonlinear problems. We show how to treat sigma point placement in a UKF as a learning problem in a model based view. We demonstrate that learning to place the sigma points correctly from data can make sigma point collapse much less likely. Learning can result in a significant increase in predictive performance over default settings of the parameters in the UKF and other filters designed to avoid the problems of the UKF, such as the GP-ADF. At the same time, we maintain a lower computational complexity than the other methods. We call our method UKF-L. ©2010 IEEE.

Model based learning of sigma points in unscented Kalman filtering

The unscented Kalman filter (UKF) is a widely used method in control and time series applications. The UKF suffers from arbitrary parameters necessary for sigma point placement, potentially causing it to perform poorly in nonlinear problems. We show how to treat sigma point placement in a UKF as a learning problem in a model based view. We demonstrate that learning to place the sigma points correctly from data can make sigma point collapse much less likely. Learning can result in a significant increase in predictive performance over default settings of the parameters in the UKF and other filters designed to avoid the problems of the UKF, such as the GP-ADF. At the same time, we maintain a lower computational complexity than the other methods. We call our method UKF-L. © 2011 Elsevier B.V.

Minimum sensor duty cycle with guaranteed estimator performance

In this paper, we consider Kalman filtering over a network and construct the optimal sensor data scheduling schemes which minimize the sensor duty cycle and guarantee a bounded error or a bounded average error at the remote estimator. Depending on the computation capability of the sensor, we can either give a closed-form expression of the minimum sensor duty cycle or provide tight lower and upper bounds of it. Examples are provided throughout the paper to demonstrate the results. © 2012 IEEE.

Measurement-Integrated simulations and Kalman filter applied to a co-flowing jet

This paper deals with the experimental evaluation of a flow analysis system based on the integration between an under-resolved Navier-Stokes simulation and experimental measurements with the mechanism of feedback (referred to as Measurement-Integrated simulation), applied to the case of a planar turbulent co-flowing jet. The experiments are performed with inner-to-outer-jet velocity ratio around 2 and the Reynolds number based on the inner-jet heights about 10000. The measurement system is a high-speed PIV, which provides time-resolved data of the flow-field, on a field of view which extends to 20 jet heights downstream the jet outlet. The experimental data can thus be used both for providing the feedback data for the simulations and for validation of the MI-simulations over a wide region. The effect of reduced data-rate and spatial extent of the feedback (i.e. measurements are not available at each simulation time-step or discretization point) was investigated. At first simulations were run with full information in order to obtain an upper limit of the MI-simulations performance. The results show the potential of this methodology of reproducing first and second order statistics of the turbulent flow with good accuracy. Then, to deal with the reduced data different feedback strategies were tested. It was found that for small data-rate reduction the results are basically equivalent to the case of full-information feedback but as the feedback data-rate is reduced further the error increases and tend to be localized in regions of high turbulent activity. Moreover, it is found that the spatial distribution of the error looks qualitatively different for different feedback strategies. Feedback gain distributions calculated by optimal control theory are presented and proposed as a mean to make it possible to perform MI-simulations based on localized measurements only. So far, we have not been able to low error between measurements and simulations by using these gain distributions.

Single camera pose estimation using Bayesian filtering and Kinect motion priors

Traditional approaches to upper body pose estimation using monocular vision rely on complex body models and a large variety of geometric constraints. We argue that this is not ideal and somewhat inelegant as it results in large processing burdens, and instead attempt to incorporate these constraints through priors obtained directly from training data. A prior distribution covering the probability of a human pose occurring is used to incorporate likely human poses. This distribution is obtained offline, by fitting a Gaussian mixture model to a large dataset of recorded human body poses, tracked using a Kinect sensor. We combine this prior information with a random walk transition model to obtain an upper body model, suitable for use within a recursive Bayesian filtering framework. Our model can be viewed as a mixture of discrete Ornstein-Uhlenbeck processes, in that states behave as random walks, but drift towards a set of typically observed poses. This model is combined with measurements of the human head and hand positions, using recursive Bayesian estimation to incorporate temporal information. Measurements are obtained using face detection and a simple skin colour hand detector, trained using the detected face. The suggested model is designed with analytical tractability in mind and we show that the pose tracking can be Rao-Blackwellised using the mixture Kalman filter, allowing for computational efficiency while still incorporating bio-mechanical properties of the upper body. In addition, the use of the proposed upper body model allows reliable three-dimensional pose estimates to be obtained indirectly for a number of joints that are often difficult to detect using traditional object recognition strategies. Comparisons with Kinect sensor results and the state of the art in 2D pose estimation highlight the efficacy of the proposed approach.

Monte Carlo filtering and smoothing with application to time-varying spectral estimation

We develop methods for performing filtering and smoothing in non-linear non-Gaussian dynamical models. The methods rely on a particle cloud representation of the filtering distribution which evolves through time using importance sampling and resampling ideas. In particular, novel techniques are presented for generation of random realisations from the joint smoothing distribution and for MAP estimation of the state sequence. Realisations of the smoothing distribution are generated in a forward-backward procedure, while the MAP estimation procedure can be performed in a single forward pass of the Viterbi algorithm applied to a discretised version of the state space. An application to spectral estimation for time-varying autoregressions is described.

Filtering via approximate Bayesian computation

Approximate Bayesian computation (ABC) has become a popular technique to facilitate Bayesian inference from complex models. In this article we present an ABC approximation designed to perform biased filtering for a Hidden Markov Model when the likelihood function is intractable. We use a sequential Monte Carlo (SMC) algorithm to both fit and sample from our ABC approximation of the target probability density. This approach is shown to, empirically, be more accurate w.r.t.~the original filter than competing methods. The theoretical bias of our method is investigated; it is shown that the bias goes to zero at the expense of increased computational effort. Our approach is illustrated on a constrained sequential lasso for portfolio allocation to 15 constituents of the FTSE 100 share index.