arXiv subjects

Robert Piché

Publications and source records attributed to Robert Piché.

11 recordsLinked to original sources

arXiv2019

On Computational Complexity Reduction Methods for Kalman Filter Extensions

The Kalman filter and its extensions are used in a vast number of aerospace and navigation applications for nonlinear state estimation of time series. In the literature, different approaches have been proposed to exploit the structure of the state and measurement models to reduce the computational demand of the algorithms. In this tutorial, we survey existing code optimization methods and present them using unified notation that allows them to be used with various Kalman filter extensions. We develop the optimization methods to cover a wider range of models, show how different structural optimizations can be combined, and present new applications for the existing optimizations. Furthermore, we present an example that shows that the exploitation of the structure of the problem can lead to improved estimation accuracy while reducing the computational load. This tutorial is intended for persons who are familiar with Kalman filtering and want to get insights for reducing the computational demand of different Kalman filter extensions.

eess.SY↗

arXiv2018

Skew-t Filter and Smoother with Improved Covariance Matrix Approximation

Filtering and smoothing algorithms for linear discrete-time state-space models with skew-t-distributed measurement noise are proposed. The algorithms use a variational Bayes based posterior approximation with coupled location and skewness variables to reduce the error caused by the variational approximation. Although the variational update is done suboptimally using an expectation propagation algorithm, our simulations show that the proposed method gives a more accurate approximation of the posterior covariance matrix than an earlier proposed variational algorithm. Consequently, the novel filter and smoother outperform the earlier proposed robust filter and smoother and other existing low-complexity alternatives in accuracy and speed. We present both simulations and tests based on real-world navigation data, in particular GPS data in an urban area, to demonstrate the performance of the novel methods. Moreover, the extension of the proposed algorithms to cover the case where the distribution of the measurement noise is multivariate skew-$t$ is outlined. Finally, the paper presents a study of theoretical performance bounds for the proposed algorithms.

eess.SY↗

arXiv2018

Damped Posterior Linearization Filter

The iterated posterior linearization filter (IPLF) is an algorithm for Bayesian state estimation that performs the measurement update using iterative statistical regression. The main result behind IPLF is that the posterior approximation is more accurate when the statistical regression of measurement function is done in the posterior instead of the prior as is done in non-iterative Kalman filter extensions. In IPLF, each iteration in principle gives a better posterior estimate to obtain a better statistical regression and more accurate posterior estimate in the next iteration. However, IPLF may diverge. IPLF's fixed- points are not described as solutions to an optimization problem, which makes it challenging to improve its convergence properties. In this letter, we introduce a double-loop version of IPLF, where the inner loop computes the posterior mean using an optimization algorithm. Simulation results are presented to show that the proposed algorithm has better convergence than IPLF and its accuracy is similar to or better than other state-of-the-art algorithms.

math.OC↗

arXiv2017

3D angle-of-arrival positioning using von Mises-Fisher distribution

We propose modeling an angle-of-arrival (AOA) positioning measurement as a von Mises-Fisher (VMF) distributed unit vector instead of the conventional normally distributed azimuth and elevation measurements. Describing the 2-dimensional AOA measurement with three numbers removes discontinuities and reduces nonlinearity at the poles of the azimuth-elevation coordinate system. Our computer simulations show that the proposed VMF measurement noise model based filters outperform the normal distribution based algorithms in accuracy in a scenario where close-to-pole measurements occur frequently.

eess.SY↗

arXiv2016

Kullback-Leibler Divergence Approach to Partitioned Update Kalman Filter

Kalman filtering is a widely used framework for Bayesian estimation. The partitioned update Kalman filter applies a Kalman filter update in parts so that the most linear parts of measurements are applied first. In this paper, we generalize partitioned update Kalman filter, which requires the use oft the second order extended Kalman filter, so that it can be used with any Kalman filter extension. To do so, we use a Kullback-Leibler divergence approach to measure the nonlinearity of the measurement, which is theoretically more sound than the nonlinearity measure used in the original partitioned update Kalman filter. Results show that the use of the proposed partitioned update filter improves the estimation accuracy.

math.OC↗

arXiv2016

Partitioned Update Kalman Filter

In this paper we present a new Kalman filter extension for state update called Partitioned Update Kalman Filter (PUKF). PUKF updates the state using multidimensional measurements in parts. PUKF evaluates the nonlinearity of the measurement function within a Gaussian prior by comparing the effect of the 2nd order term on the Gaussian measurement noise. A linear transformation is applied to measurements to minimize the nonlinearity of a part of the measurement. The measurement update is then applied using only the part of the measurement that has low nonlinearity and the process is then repeated for the updated state using the remaining part of the transformed measurement until the whole measurement has been used. PUKF does the linearizations numerically and no analytical differentiation is required. Results show that when the measurement geometry allows effective partitioning, the proposed algorithm improves estimation accuracy and produces accurate covariance estimates.

math.OC↗

arXiv2015

Robust Inference for State-Space Models with Skewed Measurement Noise

Filtering and smoothing algorithms for linear discrete-time state-space models with skewed and heavy-tailed measurement noise are presented. The algorithms use a variational Bayes approximation of the posterior distribution of models that have normal prior and skew-t-distributed measurement noise. The proposed filter and smoother are compared with conventional low-complexity alternatives in a simulated pseudorange positioning scenario. In the simulations the proposed methods achieve better accuracy than the alternative methods, the computational complexity of the filter being roughly 5 to 10 times that of the Kalman filter.

eess.SY↗

arXiv2014

Bayesian Hierarchical Model of Total Variation Regularisation for Image Deblurring

A Bayesian hierarchical model for total variation regularisation is presented in this paper. All the parameters of an inverse problem, including the "regularisation parameter", are estimated simultaneously from the data in the model. The model is based on the characterisation of the Laplace density prior as a scale mixture of Gaussians. With different priors on the mixture variable, other total variation like regularisations e.g. a prior that is related to t-distribution, are also obtained. An approximation of the resulting posterior mean is found using a variational Bayes method. In addition, an iterative alternating sequential algorithm for computing the maximum a posteriori estimate is presented. The methods are illustrated with examples of image deblurring. Results show that the proposed model can be used for automatic edge-preserving inversion in the case of image deblurring. Despite promising results, some difficulties with the model were encountered and are subject to future work.

math.NA↗

arXiv2014

Gaussian filtering and variational approximations for Bayesian smoothing in continuous-discrete stochastic dynamic systems

The Bayesian smoothing equations are generally intractable for systems described by nonlinear stochastic differential equations and discrete-time measurements. Gaussian approximations are a computationally efficient way to approximate the true smoothing distribution. In this work, we present a comparison between two Gaussian approximation methods. The Gaussian filtering based Gaussian smoother uses a Gaussian approximation for the filtering distribution to form an approximation for the smoothing distribution. The variational Gaussian smoother is based on minimizing the Kullback-Leibler divergence of the approximate smoothing distribution with respect to the true distribution. The results suggest that for highly nonlinear systems, the variational Gaussian smoother can be used to iteratively improve the Gaussian filtering based smoothing solution. We also present linearization and sigma-point methods to approximate the intractable Gaussian expectations in the Variational Gaussian smoothing equations. In addition, we extend the variational Gaussian smoother for certain class of systems with singular diffusion matrix.

math.DS↗

arXiv2012

On the Parametric Instability Caused by Step Size Variation in Runge-Kutta-Nyström Methods

The parametric instability arising when ordinary differential equations (ODEs) are numerically integrated with Runge-Kutta-Nyström (RKN) methods with varying step sizes is investigated. It is shown that when linear constant coefficient ODEs are integrated with RKN methods that are based on A-stable Runge-Kutta methods, the solution is nonincreasing in some norm for all positive step sizes, constant or varying. Perturbation methods are used to quantify the critical step sizes associated with parametric instability.

math.NA↗

arXiv2012

Stock Price Dynamics and Option Valuations under Volatility Feedback Effect

According to the volatility feedback effect, an unexpected increase in squared volatility leads to an immediate decline in the price-dividend ratio. In this paper, we consider the properties of stock price dynamics and option valuations under the volatility feedback effect by modeling the joint dynamics of stock price, dividends, and volatility in continuous time. Most importantly, our model predicts the negative effect of an increase in squared return volatility on the value of deep-in-the-money call options and, furthermore, attempts to explain the volatility puzzle. We theoretically demonstrate a mechanism by which the market price of diffusion return risk, or an equity risk-premium, affects option prices and empirically illustrate how to identify that mechanism using forward-looking information on option contracts. Our theoretical and empirical results support the relevance of the volatility feedback effect. Overall, the results indicate that the prevailing practice of ignoring the time-varying dividend yield in option pricing can lead to oversimplification of the stock market dynamics.

q-fin.PR↗