Number of the records: 1
Two Learning Approaches to Maze Exploration: Case Study with E-puck Mobile Robots
- 1.0317083 - ÚI 2009 RIV US eng C - Conference Paper (international conference)
Neruda, Roman - Slušný, Stanislav - Vidnerová, Petra
Two Learning Approaches to Maze Exploration: Case Study with E-puck Mobile Robots.
[Dva přístupy k učení úlohy prohledávání bludiště: příkladová studie s mobilními roboty E-puck.]
World Congress on Engineering and Computer Science. Hong Kong: Newswood Limited, 2008, s. 655-660. Lecture Notes in Engineering and Computer Science. ISBN 978-988-98671-0-2.
[WCECS 2008. World Congress on Engineering and Computer Science. San Francisco (US), 22.10.2008-24.10.2008]
R&D Projects: GA MŠMT(CZ) 1M0567
Institutional research plan: CEZ:AV0Z10300504
Keywords : evolutionary robotics * reinforcement learning * maze exploration
Subject RIV: IN - Informatics, Computer Science
http://www.iaeng.org/publication/WCECS2008/WCECS2008_pp655-660.pdf
A case study with experimental mobile E-puck robots trained to explore an unknown maze is presented. Learning phase is done off-line in a software simulator, while the evaluation is made both in software simulation of different maze and with physical robots.
Je prezentována příkladová studie s mobilními roboty E-puck, které jsou trénovány na úlohu prozkoumání neznámého bludiště. Učící fáze je realizována off-line v softwarovém simulátoru, evaluace je jak softwarová, tak i s fyzickými roboty.
Permanent Link: http://hdl.handle.net/11104/0166827
Number of the records: 1