Detail publikace

Walking Gait of Four-Legged Robot Obtained Throug Q Learning

BŘEZINA, T., HOUŠKA, P., SINGULE, V.

Originální název

Walking Gait of Four-Legged Robot Obtained Throug Q Learning

Český název

Walking Gait of Four-Legged Robot Obtained Throug Q Learning

Typ

článek ve sborníku ve WoS nebo Scopus

Jazyk

cs

Originální abstrakt

The possible method of walking policy obtaining of four-legged robot through Q-learning is discussed in the contribution. Q-learning is implemented using architecture represented by nondeterministic state machine that defines both possible discrete states and admissible transitions between them. Discrete state is designed as indicators vector of goals achievement by single simultaneously activated instances of two basic controllers. Only simultaneous activations that guarantee static stability of robot are admissible even in the case when single activations cold not achieve its goals. The controllers attempt to achieve its goals using on-line minimization process. Q-learning sequentially improves an estimation of future benefit from usage of admissible simultaneous activations in single discrete states. Walking policy is generated through activations with the highest estimation of future benefit.

Český abstrakt

The possible method of walking policy obtaining of four-legged robot through Q-learning is discussed in the contribution. Q-learning is implemented using architecture represented by nondeterministic state machine that defines both possible discrete states and admissible transitions between them. Discrete state is designed as indicators vector of goals achievement by single simultaneously activated instances of two basic controllers. Only simultaneous activations that guarantee static stability of robot are admissible even in the case when single activations cold not achieve its goals. The controllers attempt to achieve its goals using on-line minimization process. Q-learning sequentially improves an estimation of future benefit from usage of admissible simultaneous activations in single discrete states. Walking policy is generated through activations with the highest estimation of future benefit.

Rok RIV

2003

Vydáno

12.05.2003

Nakladatel

Institute of Theoretical and Applied Mechanics, Academy of Sciences of the Czech Republic, Prague

Místo

Praha

ISBN

80-86246-18-3

Kniha

Engeneering Mechanics 2003

Číslo edice

1

Strany od

36

Strany do

37

Strany počet

2

BibTex


@inproceedings{BUT7626,
  author="Tomáš {Březina} and Pavel {Houška} and Vladislav {Singule}",
  title="Walking Gait of Four-Legged Robot Obtained Throug Q Learning",
  annote="The possible method of walking policy obtaining of four-legged robot through Q-learning is discussed in the contribution. Q-learning is implemented using architecture represented by nondeterministic state machine that defines both possible discrete states and admissible transitions between them. Discrete state is designed as indicators vector of goals achievement by single simultaneously activated instances of two basic controllers. Only simultaneous activations that guarantee static stability of robot are admissible even in the case when single activations cold not achieve its goals. The controllers attempt to achieve its goals using on-line minimization process. Q-learning sequentially improves an estimation of future benefit from usage of admissible simultaneous activations in single discrete states. Walking policy is generated through activations with the highest estimation of future benefit.",
  address="Institute of Theoretical and Applied Mechanics, Academy of Sciences of the Czech Republic, Prague",
  booktitle="Engeneering Mechanics 2003",
  chapter="7626",
  institution="Institute of Theoretical and Applied Mechanics, Academy of Sciences of the Czech Republic, Prague",
  year="2003",
  month="may",
  pages="36",
  publisher="Institute of Theoretical and Applied Mechanics, Academy of Sciences of the Czech Republic, Prague",
  type="conference paper"
}