DocumentCode
2640640
Title
Emergent behavior based on disparity of individual capability
Author
Iio, Takamasa ; Tanev, Ivan ; Shimohara, Katsunori
Author_Institution
Doshisha Univ., Kyoto
fYear
2007
fDate
17-20 Sept. 2007
Firstpage
1588
Lastpage
1592
Abstract
We present the result of our research on the emergent behavior of heterogeneous agents in multi-agent system (MAS). We extended the commonly considered MAS for the predator prey problem (PPP), initially comprising homogeneous predator agents into heterogeneous MAS featuring predator agents with diverse sensory and moving abilities. Moreover, we evolved, via genetic programming, the behavior of these agents, and discovered the emergence of a characteristic behavior that is (i) unique to the heterogeneous systems, (ii) well specialized to fit the individual morphological features of the agents, and (iii) more efficient than that of the homogeneous agents.
Keywords
genetic algorithms; multi-agent systems; genetic programming; heterogeneous MAS; homogeneous predator agents; multiagent system; predator prey problem; Application software; Artificial intelligence; Computer science; Distributed computing; Genetic programming; Information systems; Intelligent robots; Knowledge engineering; Multiagent systems; Problem-solving; Credit Assignment Problem; Emergent Behavior; Heterogeneous Agents; Multi-Agent System;
fLanguage
English
Publisher
ieee
Conference_Titel
SICE, 2007 Annual Conference
Conference_Location
Takamatsu
Print_ISBN
978-4-907764-27-2
Electronic_ISBN
978-4-907764-27-2
Type
conf
DOI
10.1109/SICE.2007.4421236
Filename
4421236
Link To Document