Autonomous motion recognition by combining reinforcement learning and hidden Markov model

Systems and Computers in Japan

Volumn 37, Issue 14, 2006, Pages 34-43

  Morooka, Ken'ichi ^a   Hamamoto, Kazuhisa ^a   Nagahashi, Hiroshi ^a  

^a TOKYO INSTITUTE OF TECHNOLOGY   (Japan)

Author keywords

Hidden Markov model; Motion generation; Motion recognition; Reinforcement learning

Indexed keywords

DATA REDUCTION; LEARNING SYSTEMS; MARKOV PROCESSES; MATHEMATICAL MODELS; RELIABILITY THEORY; SAMPLING;

HIDDEN MARKOV MODELS (HMM); MOTION GENERATION; MOTION RECOGNITION; REINFORCEMENT LEARNING;

MOTION ESTIMATION;

EID: 33751193882     PISSN: 08821666     EISSN: 1520684X     Source Type: Journal    
DOI: 10.1002/scj.20586     Document Type: Article

References (16)

1. Asada M. From motor learning to behavior learning and cognitive development. J Robotics Soc Japan 2004;22:150-155.
2. Hasegawa Y. Construction robots, themes for practical use. J Robotics Soc Japan 2003;21:232-235.
3. Ho M, Yamada Y, Umetani Y. An HMM-based temporal difference learning with model-updating capability for visual tracking of human communicational behaviors. IEEE Int Conf on Automatic Face and Gesture Recognition, p 163-168, 2002.
4. Inamura T, Nakamura Y, Toshima I, Ezaki H. An integrated model of imitation learning and symbol emergence based on mimesis theory. J Robotics Soc Japan 2004;22:108-115.
5. Kimura M, Kobayashi S. Actor-critic algorithms that use aptitude history as actor: Reinforcement learning based on incomplete value-functions. Japan Soc Artif Intell 2000; 15:267-275.
6. Morimoto J, Doya K. Acquisition of standup motion by an actual robot using hierarchical reinforcement learning. 5th Robotics Symposia, p 397-402, 2000.
7. Nakaoka S, Nakazawa A, Yokoi K, Hirukawa H, Ikeuchi K. Generating whole body motions for biped humanoid robot from captured human dances. Proc International Conference on Robotics and Automation, p 3905-3910, 2003.
8. Ogawahara K, Takamatsu J, Kimura H, Ikeuchi K. Manual modeling technique based on integration of multiple teaching motions. Meeting on Image Recognition and Understanding (MIRU2002), Vol. I, p 147-152.
9. Saito Y. Welfare and care assisting robot. J Robotics Soc Japan 2003;21:340-345.
10. Samejima K, Katagiri K, Doya K, Kawahito M. Mimetic learning and symbolization of motor patterns by competing modules. Trans IEICE 2002;J85-D-II:90-100.
11. Sato M, Ishii S. On-line EM algorithm for the normalized Gaussian network. Neural Computation 2000; 12:407-432.
12. Shibata K, Maehara S, Sugisaka M, Itoh H. Proceedings of the 12th Autonomous Distributed System Symposium, p 133-138, 2001.
13. Sutton RS, Barto AG. Reinforcement learning. MIT Press; 1998.
14. Takahashi Y, Asada M. Behavior acquisition by multi-layered reinforcement learning. J Robotics Soc Japan 2000; 18:1040-1046.
15. Takahashi Y, Asada M. State-action space construction for multi-layered learning system. J Robotics Soc Japan 2003;21:164-171.
16. Umetani Y, Yoshida K. Space robot. J Robot Soc Japan 2003;21:229-231.