Nonlinear State Estimation Based on Deterministic Sampling
Speaker: Uwe D. Hanebeck, Karlsruhe Institute of Technology, Germany
Department: Electrical Engineering
Location: Engineering Quadrangle B205
Date/Time: Tuesday, December 10, 2013, 4:30 p.m. - 5:30 p.m.
This talk is about designing nonlinear filters for estimating the state of nonlinear stochastic systems based on approximating prior densities by deterministic samples called Dirac mixture approximation. As an example, a new Gaussian filter is derived that relies on two main ingredients: i) the progressive inclusion of the measurement information and ii) a tight coupling between a Gaussian density and its deterministic Dirac mixture approximation. No second Gaussian assumption for the joint density of state and measurement is required, so that the performance is much better than that of Linear Regression Kalman Filters (LRKFs), which heavily rely on this assumption. It can be used as a plug-in replacement for standard Gaussian filters such as the Unscented Kalman Filter (UKF).