Filling the gap between low frequency measurements with their estimates
2014 (English)In: Robotics and Automation (ICRA), 2014 IEEE International Conference on, IEEE conference proceedings, 2014, 175-180 p.Conference paper (Refereed)
The use of redundant sensors brings a rich diver-sity of information, nevertheless fusing different sensors thatrun at vastly different frequencies into a proper estimate isstill a challenging sensor fusion problem. Instead of using thesize-varying measurements and thereby the size-varying filtersduring each sampling period, we propose to find a substitute ofthe unavailable low frequency measurements such that we canavoid using different sampling frequencies in one filter. In thegap between the sampling of two low frequency measurements,the use of these substitutes produces smoother estimates. In boththe proof of concept simulation and the localization experimentperformed on an indoor soccer robot, our proposed approachexhibits an improved performance compared to the size-varyingKalman filter methods.
Place, publisher, year, edition, pages
IEEE conference proceedings, 2014. 175-180 p.
Research subject Computer Science
IdentifiersURN: urn:nbn:se:kth:diva-165761DOI: 10.1109/ICRA.2014.6906606OAI: oai:DiVA.org:kth-165761DiVA: diva2:808746
Robotics and Automation (ICRA), 2014 IEEE International Conference on, Hong Kong,May 31 2014-June 7 2014
FunderSwedish Research CouncilSwedish Foundation for Strategic Research EU, FP7, Seventh Framework Programme, FP7-ICT-288533
QC 201505072015-04-292015-04-292015-05-07Bibliographically approved