Logo des Repositoriums
 

Sensing and rating different environmental states: states: A basic approach exemplified on moving in unknown terrain

dc.contributor.authorDeutscher, Manfred
dc.contributor.authorKatz, Marcel
dc.contributor.authorKrüger, Sven
dc.contributor.editorCremers, Armin B.
dc.contributor.editorManthey, Rainer
dc.contributor.editorMartini, Peter
dc.contributor.editorSteinhage, Volker
dc.date.accessioned2019-10-11T07:58:03Z
dc.date.available2019-10-11T07:58:03Z
dc.date.issued2005
dc.description.abstractIt is difficult to describe in an analytical way all dynamical relationships of the ground and a robot or a human. Unexpected environmental changes complicate a safe walking.Therefore it is obvious to develop a method what allows the evaluation of acoustic emissions depending on acting forces in the past and consecutive steps during walking. The analytical evaluation of the relationship between robot body and the environmental reactions and a subjective interpretation delivers bad or good steps of these anticipations. That knowledge can be "leaved" for further explorations. The fusion of our data and evaluation of these relationships were done by our Support Vector Machine (SVM) a fast statistical classifier. The recognizable relationships between the empirical information and that of the SVM can help to generate safe movements.en
dc.identifier.isbn3-88579-397-0
dc.identifier.pissn1617-5468
dc.identifier.urihttps://dl.gi.de/handle/20.500.12116/28220
dc.language.isoen
dc.publisherGesellschaft für Informatik e.V.
dc.relation.ispartofInformatik 2005 – Informatik Live! Band 2
dc.relation.ispartofseriesLecture Notes in Informatics (LNI) - Proceedings, Volume P-68
dc.titleSensing and rating different environmental states: states: A basic approach exemplified on moving in unknown terrainen
dc.typeText/Conference Paper
gi.citation.endPage306
gi.citation.publisherPlaceBonn
gi.citation.startPage302
gi.conference.date19. bis 22. September 2005
gi.conference.locationBonn
gi.conference.sessiontitleRegular Research Papers

Dateien

Originalbündel
1 - 1 von 1
Lade...
Vorschaubild
Name:
GI-Proceedings.68-63.pdf
Größe:
249 KB
Format:
Adobe Portable Document Format