DocumentCode
2007246
Title
Simulation Experiments with Bio-inspired Algorithms for Odor Source Localization in Laminar Wind Flow
Author
Lochmatter, Thomas ; Martinoli, Alcherio
Author_Institution
Distrib. Intell. Syst. & Algorithms Lab., Ecole Polytech. Fed. de Lausanne, Lausanne
fYear
2008
fDate
11-13 Dec. 2008
Firstpage
437
Lastpage
443
Abstract
We compare three bio-inspired odor source localization algorithm (casting, surge-spiral and surge-cast) for environments with a main wind flow in simulation. The wind flow is laminar and the simulation setup similar to the setup in the wind tunnel in which we have carried out similar experiments with real robots. The algorithms are compared in terms of success rate and distance overhead when tracking the plume up to the source. We conclude that the algorithms based on upwind surge yield significantly better performance than pure casting.
Keywords
electronic noses; flow simulation; laminar flow; wind tunnels; bio-inspired algorithms; laminar wind flow; odor source localization; plume; wind tunnel; Casting; Chemical sensors; Intelligent structures; Intelligent systems; Learning systems; Machine learning; Machine learning algorithms; Robot sensing systems; Service robots; Surges; Odor source localization; mobile robot; odor sensor; robotic simulator; webots; wind sensor;
fLanguage
English
Publisher
ieee
Conference_Titel
Machine Learning and Applications, 2008. ICMLA '08. Seventh International Conference on
Conference_Location
San Diego, CA
Print_ISBN
978-0-7695-3495-4
Type
conf
DOI
10.1109/ICMLA.2008.128
Filename
4725010
Link To Document