DocumentCode :
1617573
Title :
The ontological status of autonomous underwater vehicle fleets
Author :
Ray, Patrick ; O´Rourke, M. ; Edwards, Dean
Author_Institution :
Center for Intell. Syst. Res., Univ. of Idaho, Moscow, ID, USA
fYear :
2009
Firstpage :
1
Lastpage :
7
Abstract :
Collective behavior differs essentially from aggregate individual behavior. Multiple agents acting collectively are capable of performing certain behaviors that cannot be reduced to the behaviors of multiple agents acting individually. When multiple agents perform behaviors, they can be acting collectively and so be characterized as one entity. Ant colonies are generally thought to behave as a single entity rather than as a mere aggregate of individuals, and human social behavior also exhibits advantages over aggregate individual behavior. Research has shown the advantages of single entity, collective behavior, known as "emergent behavior", over aggregate sum behavior. In modeling the behaviors of multiple agents, we argue that emergent behavior must be characterized collectively. Autonomous underwater vehicle (AUV) fleets are collectives of autonomous vehicles with behaviors emerging at a fleet level. For this reason, the AUV behavior designer ought to distinguish between individual behavior and collective, fleet behaviors and use this in a dynamic and bi-level approach to structuring the pursuit of mission objectives.
Keywords :
mobile robots; multi-agent systems; multi-robot systems; ontologies (artificial intelligence); optimisation; remotely operated vehicles; underwater vehicles; ant colonies; autonomous underwater vehicle fleets; behavior modeling; bi-level approach; emergent behavior; multiple agents; ontological status; Aggregates; Collaboration; Humans; Intelligent systems; Intelligent vehicles; Legged locomotion; Ontologies; Painting; Remotely operated vehicles; Underwater vehicles;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
OCEANS 2009, MTS/IEEE Biloxi - Marine Technology for Our Future: Global and Local Challenges
Conference_Location :
Biloxi, MS
Print_ISBN :
978-1-4244-4960-6
Electronic_ISBN :
978-0-933957-38-1
Type :
conf
Filename :
5422192
Link To Document :
بازگشت