This paper presents an experimental investigation about Reinforcement Learning of multiple reactive behaviors in an autonomous mobile robot. Reactive behaviors are learned concurrently or sequentially in real-time and coping with finite lifespan, limited accuracy, and finite computational resources constraints. Results provide insight into the problems and tradeoffs faced in learning composite behaviors.
Titolo: | Experiments in Sequenced Reinforcement Learning of Reactive Tasks by a Mobile Robot |
Autori: | |
Data di pubblicazione: | 2000 |
Abstract: | This paper presents an experimental investigation about Reinforcement Learning of multiple reactive behaviors in an autonomous mobile robot. Reactive behaviors are learned concurrently or sequentially in real-time and coping with finite lifespan, limited accuracy, and finite computational resources constraints. Results provide insight into the problems and tradeoffs faced in learning composite behaviors. |
Handle: | http://hdl.handle.net/11381/1451868 |
ISBN: | 1586030787 |
Appare nelle tipologie: | 4.1b Atto convegno Volume |
File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.