Sequential bayesian filtering

From Wikipedia, the free encyclopedia

Sequential Bayesian filtering is a method to estimate the real value of an observed variable that evolves in time. The method is named filtering when we estimate the current value given past observations, smoothing when estimating past value given present and past measures, and prediction when estimating a probable future value.

It is the extension of the Bayesian estimation for the case when the observed value change in time.

( should add some graphical models here, and the Bayesian program )

This is a generic concept, well known specific cases are:

• Kalman filter, for linear systems with Gaussian noise
• Particle filter, to estimate the state of nonlinear systems

The notion of Sequential Bayesian filtering is extensively used in control and robotics.

[edit] References

• A survey of probabilistic models, using the Bayesian Programming methodology as a unifying framework