Logo des Repositoriums
 
Konferenzbeitrag

Learning to control Forest Fires

Zusammenfassung

Forest fires are an important environmental problem. This paper describes a methodology for constructing an intelligent system which aims to support the human expert's decision making in fire control. The idea is based on first implementing a fire spread simulator and on searching for good decision policies by reinforcement learning (RL). RL algorithms optimize policies by letting the agents interact with the simulator and learn from their experiences. Finally, we observe different problems and propose solutions for solving them. Among these problems are storing policies for huge state spaces and coping with multiple agents which need to learn cooperative strategies.

Beschreibung

Wiering, Marco; Dorigo, Marco (1998): Learning to control Forest Fires. Umweltinformatik ’98 - Vernetzte Strukturen in Informatik, Umwelt und Wirtschaft - Computer Science for Environmental Protection ’98 - Networked Structures in Information Technology, the Environment and Business. Marburg: Metropolis. Wissensverarbeitung in Umweltanwendungen, Knowledge Processing for Environmental Applications. Bremen. 1998

Schlagwörter

Zitierform

DOI

Tags