Recursive total least squares: An alternative to using the discrete Kalman filter in robot navigation

Daniel L Boley, Erik S. Steinmetz, Karen T. Sutherland

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations

Abstract

In the robot navigation problem, noisy sensor data must be filtered to obtain the best estimate of the robot position. The discrete Kalman filter, commonly used for prediction and detection of signals in communication and control problems, has become a popular method to reduce the effect of uncertainty from the sensor data. However, in the domain of robot navigation, sensor readings are not only uncertain, but can also be relatively infrequent compared to traditional signal processing applications. In addition, a good initial estimate of location, critical for Kalman convergence, is often not available. Hence, there is a need for a filter that is capable of converging with a poor initial estimate and many fewer readings than the Kalman filter. To this end. we propose the use of a Recursive Total Least Squares Filter. This filter is easily updated to incorporate new sensor data, and in our experiments converged faster and to greater accuracy than the Kalman filter.

Original languageEnglish (US)
Title of host publicationReasoning with Uncertainty in Robotics - International Workshop, RUR 1995, Proceedings
EditorsMichiel van Lambalgen, Frans Voorbraak, Leo Dorst
PublisherSpringer Verlag
Pages221-234
Number of pages14
ISBN (Print)3540613765, 9783540613763
DOIs
StatePublished - Jan 1 1996
EventInternational Workshop on Reasoning with Uncertainty in Robotics, RUR 1995 - Amsterdam, Netherlands
Duration: Dec 4 1995Dec 6 1995

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1093
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Workshop on Reasoning with Uncertainty in Robotics, RUR 1995
CountryNetherlands
CityAmsterdam
Period12/4/9512/6/95

Fingerprint Dive into the research topics of 'Recursive total least squares: An alternative to using the discrete Kalman filter in robot navigation'. Together they form a unique fingerprint.

Cite this