A Probabilistic Perspective on Gaussian Filtering and Smoothing - Statistics > MethodologyReport as inadecuate




A Probabilistic Perspective on Gaussian Filtering and Smoothing - Statistics > Methodology - Download this document for free, or read online. Document in PDF available to download.

Abstract: We present a general probabilistic perspective on Gaussian filtering andsmoothing. This allows us to show that common approaches to Gaussianfiltering-smoothing can be distinguished solely by their methods ofcomputing-approximating the means and covariances of joint probabilities. Thisimplies that novel filters and smoothers can be derived straightforwardly byproviding methods for computing these moments. Based on this insight, we derivethe cubature Kalman smoother and propose a novel robust filtering and smoothingalgorithm based on Gibbs sampling.



Author: Marc Peter Deisenroth, Henrik Ohlsson

Source: https://arxiv.org/







Related documents