Learning to Grasp Unknown Objects Based on 3D Edge Information

Leon Bodenhagen, Dirk Kraft, Mila Popovic, Emre Baseski, Peter Eggenberger Hotz, Norbert Krüger

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

In this work we refine an initial grasping behavior based on 3D edge information by learning. Based on a set of autonomously generated evaluated grasps and relations between the semi-global 3D edges, a prediction function is learned that computes a likelihood for the success of a grasp using either an offline or an online learning scheme. Both methods are implemented using a hybrid artificial neural network containing standard nodes with a sigmoid activation function and nodes with a radial basis function. We show that a significant performance improvement can be achieved.
Original languageEnglish
Title of host publicationIEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA2009)
PublisherIEEE
Publication date2010
Pages421 - 428
Publication statusPublished - 2010
EventThe 8th IEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA2009) - DAEJEON, Korea, Republic of
Duration: 15. Dec 200918. Dec 2009

Conference

ConferenceThe 8th IEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA2009)
Country/TerritoryKorea, Republic of
CityDAEJEON
Period15/12/200918/12/2009

Fingerprint

Dive into the research topics of 'Learning to Grasp Unknown Objects Based on 3D Edge Information'. Together they form a unique fingerprint.

Cite this