Probabilistic Modeling of Human Dynamics for Intention Inference

File Description SizeFormat 
p55.pdfPublished version1.87 MBAdobe PDFDownload
Title: Probabilistic Modeling of Human Dynamics for Intention Inference
Author(s): Wang, Z
Deisenroth, MP
Amor, HB
Vogt, D
Schölkopf, B
Peters, J
Item Type: Conference Paper
Abstract: © 2013 Massachusetts Institute of Technology.Inference of human intention may be an essential step towards understanding human actions and is hence important for realizing efficient human-robot interaction. In this paper, we propose the Intention-Driven Dynamics Model (IDDM), a latent variable model for inferring unknown human intentions. We train the model based on observed human movements/actions. We introduce an efficient approximate inference algorithm to infer the humans intention from an ongoing movement. We verify the feasibility of the IDDM in two scenarios, i.e., target inference in robot table tennis and action recognition for interactive humanoid robots. In both tasks, the IDDM achieves substantial improvements over state-of-The-Art regression and classification.
Publication Date: 31-Jul-2012
URI: http://hdl.handle.net/10044/1/12199
ISBN: 978-981-07-3937-9
Publisher: MIT Press
Journal / Book Title: Proceedings of Robotics: Science & Systems (RSS 2012)
Copyright Statement: © 2012 The Authors
Conference Name: Robotics Science & Systems VIII
Publisher URL: http://www.roboticsproceedings.org/rss08/p55.html
Start Date: 2012-07-09
Finish Date: 2012-07-13
Conference Place: Sdyney, Austrialia
Appears in Collections:Computing



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

Creative Commons