Hierarchies of Coupled Inverse and Forward Models for Abstraction in Robot Action Planning, Recognition and Imitation

File Description SizeFormat 
aisb05.pdfAccepted version715.01 kBAdobe PDFView/Open
Title: Hierarchies of Coupled Inverse and Forward Models for Abstraction in Robot Action Planning, Recognition and Imitation
Authors: Johnson, M
Demiris, Y
Item Type: Conference Paper
Abstract: Coupling internal inverse and forward models gives rise to on-line simulation processes that may be used as a common computational substrate for action execution, planning, recognition, imitation and learning. In this paper, multiple coupled internal inverse and forward models are arranged in a hierarchical fashion, with each level of the hierarchy interacting with other levels through top-down and bottom-up processes. Through experiments involving imitation of a human demonstrator performing object manipulation tasks, this architecture is shown to equip a robot with a multi-level motor abstraction capability. This is then used to solve the correspondence problem in action recognition. The architecture is inspired by biological evidence.
Issue Date: 30-Apr-2005
URI: http://hdl.handle.net/10044/1/12622
Publisher: AISB
Start Page: 69
End Page: 76
Journal / Book Title: Proceedings of the Third International Symposium on Imitation in Animals and Artifacts
Copyright Statement: © 2005 The Authors
Conference Name: International Symposium on Imitation in Animals and Artifacts
Conference Location: Hatfield, UK
Publication Status: Published
Publisher URL: http://www.aisb.org.uk/asibpublications/convention-proceedings
Start Date: 2005-04-12
Finish Date: 2005-04-15
Conference Place: Hatfield, UK
Appears in Collections:Electrical and Electronic Engineering



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx