Comparison of nonlinear filtering techniques

Filtering;state estimation;particle filtering;Kalman filter;feedback particle filter

Ghiotto, Shane ; Mehta ; Prashant G.

关键词: Filtering;    state estimation;    particle filtering;    Kalman filter;    feedback particle filter;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/49437/Shane_Ghiotto.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

In a recent work it is shown that importance sampling can be avoided in theparticle filter through an innovation structure inspired by traditional nonlinearfiltering combined with optimal control formalisms. The resulting algorithm isreferred to as feedback particle filter.The purpose of this thesis is to provide a comparative study of the feedbackparticle filter (FPF). Two types of comparisons are discussed: i) with the extendedKalman filter, and ii) with the conventional resampling-based particle filters. Thecomparison with Kalman filter is used to highlight the feedback structure of theFPF. Also computational cost estimates are discussed, in terms of number of op-erations relative to EKF. Comparison with the conventional particle filtering ap-proaches is based on a numerical example taken from the survey article on thetopic of nonlinear filtering. Comparisons are provided for both computationalcost and accuracy.