Bayesian Filtering and Smoothing

@inproceedings{Srkk2013BayesianFA,
  title={Bayesian Filtering and Smoothing},
  author={Simo S{\"a}rkk{\"a}},
  booktitle={Institute of Mathematical Statistics textbooks},
  year={2013}
}
  • S. Särkkä
  • Published in
    Institute of Mathematical…
    2013
  • Computer Science
Filtering and smoothing methods are used to produce an accurate estimate of the state of a time-varying system based on multiple observational inputs (data). Interest in these methods has exploded in recent years, with numerous applications emerging in fields such as navigation, aerospace engineering, telecommunications, and medicine. This compact, informal introduction for graduate students and advanced undergraduates presents the current state-of-the-art filtering and smoothing methods in a… 
Robust Bayesian Filtering and Smoothing Using Student's t Distribution
TLDR
The use of Student's t distribution is described to develop robust, scalable, and simple filtering and smoothing algorithms that closely resemble the Kalman filter and the Rauch-Tung-Striebel smoother except for a nonlinear measurement-dependent matrix update.
Sigma-Point Filtering and Smoothing Based Parameter Estimation in Nonlinear Dynamic Systems
TLDR
This work considers approximate maximum likelihood parameter estimation in nonlinear state-space models and discusses both direct optimization of the likelihood and expectation--maximization (EM), and focuses on using Gaussian filtering and smoothing algorithms that employ sigma-points to approximate the required integrals.
Nonlinear Bayesian estimation: from Kalman filtering to a broader horizon
TLDR
A systematic introduction to the Bayesian state estimation framework is offered and various Kalman filtering U+0028 KF U-0029 techniques are reviewed, progressively from the standard KF for linear systems to extended KF, unscented KF and ensemble KFFor nonlinear systems.
Recursive Bayesian Filtering Through a Mixture of Gaussian and Discrete Particles
TLDR
A new type of filter is proposed where particles in addition to a (linearized) Gaussian component are tracked, which can be seen as a parallel solution to the estimation problem, each component can be separately filtered and constituent outputs summed up to form the filtering distribution.
Bayesian quadrature in nonlinear filtering
  • Jakub Prüher, M. Simandl
  • Mathematics
    2015 12th International Conference on Informatics in Control, Automation and Robotics (ICINCO)
  • 2015
TLDR
This paper improves estimation of covariances in quadrature-based filtering algorithms by taking into account the integral variance, and applies the proposed modifications to the Gauss-Hermite Kalman filter and the unscented Kalmanfilter algorithms.
Kalman filters as the steady-state solution of gradient descent on variational free energy
TLDR
This work presents a straightforward derivation of Kalman filters consistent with active inference via a variational treatment of free energy minimisation in terms of gradient descent, offering a more direct link between models of neural dynamics as gradient descent and standard accounts of perception and decision making based on probabilistic inference.
The Level Set Kalman Filter for State Estimation of Continuous-Discrete Systems
TLDR
The LSKF improves the time-update step compared to existing methods, such as the continuous-discrete cubature Kalman filter (CD-CKF), by reformulating the underlying Fokker-Planck equation as an ordinary differential equation for the Gaussian, thereby avoiding the need for the explicit expression of the higher derivatives.
Approximate Inference and Learning of State Space Models With Laplace Noise
TLDR
This paper presents a new approximate inference algorithm for state space models with Laplace-distributed multivariate data that is robust to a wide range of non-Gaussian noise, and presents a maximum posterior expectation-maximization (EM) algorithm that provides better model estimation than existing methods for the Gaussian model.
Accurate State Estimation in Continuous-Discrete Stochastic State-Space Systems With Nonlinear or Nondifferentiable Observations
TLDR
A novel method of nonlinear Kalman filtering, which unifies the best features of the accurate continuous-discrete extended and cubature Kalman filters, which is particularly effective for continuous-Discrete stochastic systems with nonlinear and/or nondifferentiable observations.
Major development under Gaussian filtering since unscented Kalman filter
TLDR
Significant developments made under Gaussian filtering since the proposition of UKF are reviewed, particularly focused on advancing the numerical approximation methods; modifying the conventional Gaussian approach to further improve the filtering performance; and constrained filtering to address the problem of discrete-time formulation of process dynamics.
...
...

References

SHOWING 1-10 OF 157 REFERENCES
Recursive Bayesian inference on stochastic differential equations
TLDR
The main contributions of this thesis are to show how the recently developed discrete-time unscented Kalman filter, particle filter, and the corresponding smoothers can be applied in the continuous-discrete setting.
Mixture Kalman filters
In treating dynamic systems, sequential Monte Carlo methods use discrete samples to represent a complicated probability distribution and use rejection sampling, importance sampling and weighted
Kalman Filtering: Theory and Practice Using MATLAB
TLDR
Kalman Filtering: Theory and Practice Using MATLAB, Fourth Edition is an ideal textbook in advanced undergraduate and beginning graduate courses in stochastic processes and Kalman filtering and appropriate for self-instruction or review by practicing engineers and scientists who want to learn more about this important topic.
On sequential Monte Carlo sampling methods for Bayesian filtering
TLDR
An overview of methods for sequential simulation from posterior distributions for discrete time dynamic models that are typically nonlinear and non-Gaussian, and how to incorporate local linearisation methods similar to those which have previously been employed in the deterministic filtering literature are shown.
Monte Carlo Smoothing for Nonlinear Time Series
We develop methods for performing smoothing computations in general state-space models. The methods rely on a particle representation of the filtering distributions, and their evolution through time
Cubature Kalman Filtering for Continuous-Discrete Systems: Theory and Simulations
TLDR
Results indicate that the CD-CKF markedly outperforms existing continuous-discrete filters in the context of radar in two respects- high dimensionality of the state and increasing degree of nonlinearity.
Novel approach to nonlinear/non-Gaussian Bayesian state estimation
TLDR
An algorithm, the bootstrap filter, is proposed for implementing recursive Bayesian filters, represented as a set of random samples, which are updated and propagated by the algorithm.
Obstacles to High-Dimensional Particle Filtering
Abstract Particle filters are ensemble-based assimilation schemes that, unlike the ensemble Kalman filter, employ a fully nonlinear and non-Gaussian analysis step to compute the probability
Smoothing algorithms for state–space models
TLDR
A generalised two-filter smoothing formula is proposed which only requires approximating probability distributions and applies to any state–space model, removing the need to make restrictive assumptions used in previous approaches to this problem.
Robust Filtering and Smoothing with Gaussian Processes
We propose a principled algorithm for robust Bayesian filtering and smoothing in nonlinear stochastic dynamic systems when both the transition function and the measurement function are described by
...
...