Particle filters for state estimation of jump Markov linear systems

By Arnaud Doucet, Neil J. Gordon, Vikram Krishnamurthy. IEEE Transactions on Signal Processing, 2001

837 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/929efe729a0c512687e3e305338c2e047f37ebc8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Particle methods for change detection, system identification, and control (2004, background): “follows one can concentrate the particles to the approximation of the lower-dimensional distribution , thus reducing the variance of the Monte Carlo estimates; see, for example, [14] and [16] for details and [12] for a detailed application.”
The importance of parameter resampling for soil moisture data assimilation into hydrologic models using the particle filter (2011, background): “In Doucet et al. (2001), an optimal choice for the importance density function is proposed: q(xt |x0:t−1,y1:t)=p(xt |x0:t−1,y1:t) (19) This pdf is optimal in the sense that it minimizes the variance of the importance weights, however, the application of equation 19 is complex from the implementation”
An Overview of Existing Methods and Recent Advances in Sequential Monte Carlo (2007, background): “In cases where a random draw cannot be made directly from the conditional distributions, it will be necessary to split the state xt into smaller sub-components [48], or to apply Metropolis-Hastings moves instead of Gibbs sampling [63], [64].”
Smoothing algorithms for state–space models (2010, methodology): “JUMP MARKOV LINEAR SYSTEM Jump Markov linear systems are widely used in several fields of signal processing including seismic signal processing, digital communications such as interference suppression in CDMA spread spectrum systems, target tracking, and de-interleaving of pulse trains[16].”
Learning Switching Models for Abnormality Detection for Autonomous Driving (2018, methodology): “The most used algorithms are the Rao-Blackwellized particle SDS (RBPF) [18] that uses a PF in the continuous state space model and a Hidden Markov model (HMM) for the discrete space; and a Markov Jump Linear Systems (MJLS) [19] that uses a combination of KF and PF.”
Particle Filter SLAM with High Dimensional Vehicle Model (2009, methodology): “Furthermore, the Rao-Blackwellized particle filter (rbpf), also denoted the marginalized particle filter (mpf) [1, 5, 7, 14, 15, 32], enables estimation of velocity, acceleration, and sensor error models by utilizing any linear Gaussian substructure in the model, which is fundamental for performance”
Particle filter theory and practice with positioning applications (2010, methodology): “In [49], the depletion problem is handled by intr oducing an additional Markov Chain Monte Carlo (MCMC) step to separate the samples.”
Sigma-point kalman filters for probabilistic inference in dynamic state-space models (2004, background): “For a proof of this, see [108] and [47].”
Particle filters for positioning, navigation, and tracking (2002, methodology): “In [15], the depletion problem is handled by introducing an additional Markov chain Monte Carlo (MCMC) step to separate the samples.”
A Survey of Fault Detection, Isolation, and Reconfiguration Methods (2010, methodology): “known as the particle filter method [98], [99].”
Monte Carlo filtering for multi target tracking and data association (2005, background): “It is, however, restricted to jump Markov linear systems (JMLS) [33].”
A Probabilistic Particle-Control Approximation of Chance-Constrained Stochastic Predictive Control (2010, background): “A considerable amount of work in estimation has shown that the “best” proposal distribution depends heavily on the particular problem being addressed, and a great deal of work has focussed on developing proposal distributions for spe-ciﬁc applications, for example, [22], [23] and [34].”
A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking (2002, methodology): “An example of an application when is a member of a finite set is a Jump–Markov linear system for tracking maneuvering targets [15], whereby the discrete modal state (defining the maneuver index) is tracked using a particle filter, and (conditioned on the maneuver index) the continuous base state is…”
Rao-Blackwellised Particle Filtering for Dynamic Bayesian Networks (2000, result): “Recent theoretical results in (Crisan, Del Moral and Lyons 1999) indicate that the restriction 5 h s @ h ¦ r f s t u ¡ is unnecessary to obtain convergence results (Doucet et al. 1999).”