Show simple item record

Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS '09)

dc.contributor.authorHoward, Matthew
dc.contributor.authorKlanke, Stefan
dc.contributor.authorGienger, Michael
dc.contributor.authorGoerick, Christian
dc.contributor.authorVijayakumar, Sethu
dc.date.accessioned2010-08-23T10:20:49Z
dc.date.available2010-08-23T10:20:49Z
dc.date.issued2009
dc.identifier.issn978-1-4244-3803-7en
dc.identifier.urihttp://hdl.handle.net/1842/3655
dc.description.abstractMany everyday human skills can be framed in terms of performing some task subject to constraints imposed by the environment. Constraints are usually unobservable and frequently change between contexts. In this paper, we present a novel approach for learning (unconstrained) control policies from movement data, where observations are recorded under different constraint settings. Our approach seamlessly integrates unconstrained and constrained observations by performing hybrid optimisation of two risk functionals. The first is a novel risk functional that makes a meaningful comparison between the estimated policy and constrained observations. The second is the standard risk, used to reduce the expected error under impoverished sets of constraints. We demonstrate our approach on systems of varying complexity, and illustrate its utility for transfer learning of a car washing task from human motion capture data.en
dc.language.isoenen
dc.subjectInformaticsen
dc.subjectRoboticsen
dc.titleRobust Constraint-consistent Learningen
dc.typeConference Paperen
dc.identifier.doi10.1109/IROS.2009.5354663en
rps.titleProc. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS '09)en
dc.extent.noOfPages8en
dc.date.updated2010-08-23T10:20:49Z


Files in this item

This item appears in the following Collection(s)

Show simple item record