Robotic Motion Coach: Effect of Motion Emphasis and Verbal Expression for Imitation Learning

  • Tetsunari Inamura
  • Keisuke Okuno
Conference paper


In this paper, a robotic motion coaching system that coaches human beings is proposed. The purpose of this robotic system is to have subjects imitate and learn imitation target motions effectively and well. By applying the Mimesis model, we integrated qualitative and quantitative evaluation of player’s imitated motion patterns, as well as introduced a method to synthesize emphatic motion patterns and to integrate verbal attention that corresponds to the degree of emphasis. Series of experiment, coaching how to perform forehand-tennis-swing, showed the feasibility of the proposing method and confirmed that emphatic motions with verbal attention improved the imitation learning of motion patterns.


  1. 1.
    N. Iwahashi. Robots that learn language: Developmental approach to human-machine conversations. In N. Sanker et al., editor, Human-Robot Interaction, pp. 95–118. I-Tech Education and Publishing, 2007.Google Scholar
  2. 2.
    W. Takano and Y. Nakamura. Integrating whole body motion primitives and natural language for humanoid robots. In IEEE-RAS Int’l Conf. on Humanoid Robots, pp. 708–713, 2008.Google Scholar
  3. 3.
    T. Inamura, H. Tanie, and Y. Nakamura. From stochastic motion generation and recognition to geometric symbol development and manipulation. In Int’l Conf. on Humanoid Robots, 2003. (CD-ROM).Google Scholar
  4. 4.
    T. Inamura and T. Shibata. Geometric proto-symbol manipulation towards language-based motion pattern synthesis and recognition. In IEEE/RSJ Int’l Conf. on Intelligent Robots and Systems, pp. 334–339, 2008.Google Scholar
  5. 5.
    K. Okuno and T. Inamura. A research on motion coaching for human beings by a robotics system that uses emphatic motion and verbal attention. In Humanoids 2010 Workshop on Humanoid Robots Learning from Human Interaction, 2010.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.National Institute of InformaticsThe Graduate University for Advanced StudiesTokyoJapan

Personalised recommendations