Comparison of two nonlinear filtering techniques - the extended Kalman filter and the feedback particle filter

In a recent work it has been shown that importance sampling can be avoided in particle filter through an innovation structure inspired by traditional nonlinear filtering combined with optimal control and mean-field game formalisms. The resulting algorithm is referred to as feedback particle filter (FPF). The purpose of this thesis is to provide a comparative study of the feedback particle filter (FPF) with the extended Kalman filter (EKF) for a scalar filtering problem which has linear signal dynamics and nonlinear observation dynamics. Different parameters of the signal model and observation model will be varied and performance of the two filtering techniques FPF, EKF will be compared.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2014

Author dc:creator

• Medarametla, Krishna Kalyan

Contributors dc:contributor

• Mehta, Prashant G.

• Extended Kalman filter
• Feedback particle filter
• Comparison
• Nonlinear filtering

Statement dc:rights

• Copyright 2014 Krishna Kalyan Medarametla

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/50584

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/50584