Advanced

Primitive-based action representation and recognition

Sanmohan ; Krüger, Volker LU ; Kragic, Danica and Kjellström, Hedvig (2011) In Advanced Robotics 25(6-7). p.871-891
Abstract

In robotics, there has been a growing interest in expressing actions as a combination of meaningful subparts commonly called motion primitives. Primitives are analogous to words in a language. Similar to words put together according to the rules of language in a sentence, primitives arranged with certain rules make an action. In this paper we investigate modeling and recognition of arm manipulation actions at different levels of complexity using primitives. Primitives are detected automatically in a sequential manner. Here, we assume no prior knowledge on primitives, but look for correlating segments across various sequences. All actions are then modeled within a single hidden Markov models whose structure is learned incrementally as... (More)

In robotics, there has been a growing interest in expressing actions as a combination of meaningful subparts commonly called motion primitives. Primitives are analogous to words in a language. Similar to words put together according to the rules of language in a sentence, primitives arranged with certain rules make an action. In this paper we investigate modeling and recognition of arm manipulation actions at different levels of complexity using primitives. Primitives are detected automatically in a sequential manner. Here, we assume no prior knowledge on primitives, but look for correlating segments across various sequences. All actions are then modeled within a single hidden Markov models whose structure is learned incrementally as new data is observed. We also generate an action grammar based on these primitives and thus link signals to symbols.

(Less)
Please use this url to cite or link to this publication:
author
; ; and
publishing date
type
Contribution to journal
publication status
published
subject
keywords
activity modeling, high-level event, imitation learning, Primitive detection
in
Advanced Robotics
volume
25
issue
6-7
pages
21 pages
publisher
Taylor & Francis
external identifiers
  • scopus:79954472581
ISSN
0169-1864
DOI
10.1163/016918611X563346
language
English
LU publication?
no
id
d822a06d-789b-4e3f-9727-050715974669
date added to LUP
2019-06-28 09:22:36
date last changed
2020-01-13 02:08:23
@article{d822a06d-789b-4e3f-9727-050715974669,
  abstract     = {<p>In robotics, there has been a growing interest in expressing actions as a combination of meaningful subparts commonly called motion primitives. Primitives are analogous to words in a language. Similar to words put together according to the rules of language in a sentence, primitives arranged with certain rules make an action. In this paper we investigate modeling and recognition of arm manipulation actions at different levels of complexity using primitives. Primitives are detected automatically in a sequential manner. Here, we assume no prior knowledge on primitives, but look for correlating segments across various sequences. All actions are then modeled within a single hidden Markov models whose structure is learned incrementally as new data is observed. We also generate an action grammar based on these primitives and thus link signals to symbols.</p>},
  author       = {Sanmohan and Krüger, Volker and Kragic, Danica and Kjellström, Hedvig},
  issn         = {0169-1864},
  language     = {eng},
  month        = {04},
  number       = {6-7},
  pages        = {871--891},
  publisher    = {Taylor & Francis},
  series       = {Advanced Robotics},
  title        = {Primitive-based action representation and recognition},
  url          = {http://dx.doi.org/10.1163/016918611X563346},
  doi          = {10.1163/016918611X563346},
  volume       = {25},
  year         = {2011},
}