Motion Imitation and Recognition using Parametric Hidden Markov Models

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

17 Citations (Scopus)

Abstract

The recognition and synthesis of parametric movements play an important role in human-robot interaction. To understand the whole purpose of an arm movement of a human agent, both its recognition (e.g., pointing or reaching) as well as its parameterization (i.e., where the agent is pointing at) are important. Only together they convey the whole meaning of an action. Similarly, to imitate a movement, the robot needs to select the proper action and parameterize it, e.g., by the relative position of the object that needs to be grasped.
We propose to utilize parametric hidden Markov models (PHMMs), which extend the classical HMMs by introducing a joint parameterization of the observation densities, to simultaneously solve the problems of action recognition, parameterization of the observed actions, and action synthesis. The proposed approach was fully implemented on a humanoid robot HOAP-3. To evaluate the approach, we focused on reaching and pointing actions. Even though the movements are very similar in appearance, our approach is able to distinguish the two movement types and discover the parameterization, and is thus enabling both, action recognition and action synthesis. Through parameterization we ensure that the synthesized movements can be applied to different configurations of the external world and are thus suitable for actions that involve the manipulation of objects.
Original languageEnglish
Title of host publication8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008
Number of pages8
PublisherIEEE
Publication date2008
ISBN (Electronic)978-1-4244-2821-2
DOIs
Publication statusPublished - 2008
EventHumanoids, IEEE-RAS International Conference on Humanoid Robots - Daejeon, Korea, Republic of
Duration: 1 Dec 20083 Dec 2008

Conference

ConferenceHumanoids, IEEE-RAS International Conference on Humanoid Robots
CountryKorea, Republic of
CityDaejeon
Period01/12/200803/12/2008

Fingerprint

Hidden Markov models
Parameterization
Robots
Human robot interaction

Keywords

  • Imitation Learning Parametric Hidden Markov Models
  • Coaching
  • Humanoid Robots,
  • Parametric Hidden Markov Models

Cite this

Herzog, D., Ude, A., & Krüger, V. (2008). Motion Imitation and Recognition using Parametric Hidden Markov Models. In 8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008 IEEE. https://doi.org/10.1109/ICHR.2008.4756002
Herzog, Dennis ; Ude, Ales ; Krüger, Volker. / Motion Imitation and Recognition using Parametric Hidden Markov Models. 8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008. IEEE, 2008.
@inproceedings{93dfe680bbb211dd887e000ea68e967b,
title = "Motion Imitation and Recognition using Parametric Hidden Markov Models",
abstract = "The recognition and synthesis of parametric movements play an important role in human-robot interaction. To understand the whole purpose of an arm movement of a human agent, both its recognition (e.g., pointing or reaching) as well as its parameterization (i.e., where the agent is pointing at) are important. Only together they convey the whole meaning of an action. Similarly, to imitate a movement, the robot needs to select the proper action and parameterize it, e.g., by the relative position of the object that needs to be grasped.We propose to utilize parametric hidden Markov models (PHMMs), which extend the classical HMMs by introducing a joint parameterization of the observation densities, to simultaneously solve the problems of action recognition, parameterization of the observed actions, and action synthesis. The proposed approach was fully implemented on a humanoid robot HOAP-3. To evaluate the approach, we focused on reaching and pointing actions. Even though the movements are very similar in appearance, our approach is able to distinguish the two movement types and discover the parameterization, and is thus enabling both, action recognition and action synthesis. Through parameterization we ensure that the synthesized movements can be applied to different configurations of the external world and are thus suitable for actions that involve the manipulation of objects.",
keywords = "Imitation Learning Parametric Hidden Markov Models, Coaching, Humanoid Robots,, Parametric Hidden Markov Models",
author = "Dennis Herzog and Ales Ude and Volker Kr{\"u}ger",
year = "2008",
doi = "10.1109/ICHR.2008.4756002",
language = "English",
booktitle = "8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008",
publisher = "IEEE",
address = "United States",

}

Herzog, D, Ude, A & Krüger, V 2008, Motion Imitation and Recognition using Parametric Hidden Markov Models. in 8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008. IEEE, Daejeon, Korea, Republic of, 01/12/2008. https://doi.org/10.1109/ICHR.2008.4756002

Motion Imitation and Recognition using Parametric Hidden Markov Models. / Herzog, Dennis; Ude, Ales; Krüger, Volker.

8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008. IEEE, 2008.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Motion Imitation and Recognition using Parametric Hidden Markov Models

AU - Herzog, Dennis

AU - Ude, Ales

AU - Krüger, Volker

PY - 2008

Y1 - 2008

N2 - The recognition and synthesis of parametric movements play an important role in human-robot interaction. To understand the whole purpose of an arm movement of a human agent, both its recognition (e.g., pointing or reaching) as well as its parameterization (i.e., where the agent is pointing at) are important. Only together they convey the whole meaning of an action. Similarly, to imitate a movement, the robot needs to select the proper action and parameterize it, e.g., by the relative position of the object that needs to be grasped.We propose to utilize parametric hidden Markov models (PHMMs), which extend the classical HMMs by introducing a joint parameterization of the observation densities, to simultaneously solve the problems of action recognition, parameterization of the observed actions, and action synthesis. The proposed approach was fully implemented on a humanoid robot HOAP-3. To evaluate the approach, we focused on reaching and pointing actions. Even though the movements are very similar in appearance, our approach is able to distinguish the two movement types and discover the parameterization, and is thus enabling both, action recognition and action synthesis. Through parameterization we ensure that the synthesized movements can be applied to different configurations of the external world and are thus suitable for actions that involve the manipulation of objects.

AB - The recognition and synthesis of parametric movements play an important role in human-robot interaction. To understand the whole purpose of an arm movement of a human agent, both its recognition (e.g., pointing or reaching) as well as its parameterization (i.e., where the agent is pointing at) are important. Only together they convey the whole meaning of an action. Similarly, to imitate a movement, the robot needs to select the proper action and parameterize it, e.g., by the relative position of the object that needs to be grasped.We propose to utilize parametric hidden Markov models (PHMMs), which extend the classical HMMs by introducing a joint parameterization of the observation densities, to simultaneously solve the problems of action recognition, parameterization of the observed actions, and action synthesis. The proposed approach was fully implemented on a humanoid robot HOAP-3. To evaluate the approach, we focused on reaching and pointing actions. Even though the movements are very similar in appearance, our approach is able to distinguish the two movement types and discover the parameterization, and is thus enabling both, action recognition and action synthesis. Through parameterization we ensure that the synthesized movements can be applied to different configurations of the external world and are thus suitable for actions that involve the manipulation of objects.

KW - Imitation Learning Parametric Hidden Markov Models

KW - Coaching

KW - Humanoid Robots,

KW - Parametric Hidden Markov Models

U2 - 10.1109/ICHR.2008.4756002

DO - 10.1109/ICHR.2008.4756002

M3 - Article in proceeding

BT - 8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008

PB - IEEE

ER -

Herzog D, Ude A, Krüger V. Motion Imitation and Recognition using Parametric Hidden Markov Models. In 8th IEEE-RAS International Conference on Humanoid Robots, 2008. Humanoids 2008. IEEE. 2008 https://doi.org/10.1109/ICHR.2008.4756002