Paper Info
Title
Reduced Dimensionality Extended Kalman Filter For Slam
Abstract
Computational complexity of the Kalman filter grows at least quadratically with the number of dimensions in the filter. This is a particular problem for applications like monocular simultaneous localization and mapping (SLAM) where it is not possible to run a single filter on a large map with many thousands of landmarks.This paper presents a method for dramatically reducing the computational complexity of the Kalman filters by reducing the dimensionality as information is acquired. We prove the validity of our method by applying it to monocular SLAM, where there is a large number of dimensions in the filter that are not subject to process noise (the landmark locations). This has the effect of reducing the cost of running a filter or allowing a single filter to process a much larger set of landmarks.Our approach also has a role to play within modern efficient sparse matrix approaches to SLAM where local information is coalesced into keyframes using Kalman filters. It also has general applicability to filtered measurement of static quantities where there are large numbers of dimensions that are not subject to process noise.
Year
DOI
Venue
2013
10.5244/C.27.117
PROCEEDINGS OF THE BRITISH MACHINE VISION CONFERENCE 2013
Field
DocType
Citations 
Computer vision,Extended Kalman filter,Fast Kalman filter,Computer science,Filter (signal processing),Curse of dimensionality,Kalman filter,Artificial intelligence,Covariance matrix,Invariant extended Kalman filter,Simultaneous localization and mapping
Conference
0
PageRank 
References 
Authors
0.34
15
2
Name
Order
Citations
PageRank
Dinesh Gamage101.01
Tom Drummond22676159.45