Improved Kalman filter initialisation using neurofuzzy estimation

Publisher Website
Google Scholar
Abstract
This paper proposes that if neurofuzzy estimators producemore accurate state estimates than those calculatedfrom the observed noisy inputs (using the known statemodel), then neurofuzzy estimates can be used to initialisethe states of Kalman and extended Kalman filters.Filters whose states have been initialised with neurofuzzyestimates should give improved performance byway of faster convergence when the filter is initialised,and when a filter is re-started after divergence.

This publication has 0 references indexed in Scilit: