Walking Gait of Four-Legged Robot Obtained Throug Q Learning

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26210%2F03%3APU36728" target="_blank" >RIV/00216305:26210/03:PU36728 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/61388998:_____/03:51030198

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

čeština

Název v původním jazyce

Walking Gait of Four-Legged Robot Obtained Throug Q Learning

Popis výsledku v původním jazyce

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 robotare 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 admissiblesimultaneous activations in single discrete states. Walking policy is generated through activations with the highest estimation of future benefit.

Název v anglickém jazyce

Walking Gait of Four-Legged Robot Obtained Throug Q Learning

Popis výsledku anglicky

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 robotare 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 admissiblesimultaneous activations in single discrete states. Walking policy is generated through activations with the highest estimation of future benefit.

Druh

D - Stať ve sborníku

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2003

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Engeneering Mechanics 2003

ISBN

80-86246-18-3

ISSN

—

e-ISSN

—

Počet stran výsledku

2

Strana od-do

36-37

Název nakladatele

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

Místo vydání

Praha

Místo konání akce

Svratka

Datum konání akce

12. 5. 2003

Typ akce podle státní příslušnosti

CST - Celostátní akce

Kód UT WoS článku

—