DocumentCode :
184963
Title :
Performance of bandit methods in acoustic relay positioning
Author :
Mei Yi Cheung ; Leighton, Jon ; Mitra, U. ; Singh, Harshavardhan ; Hover, Franz S.
Author_Institution :
Dept. of Mech. Eng., MIT, Cambridge, MA, USA
fYear :
2014
fDate :
4-6 June 2014
Firstpage :
4708
Lastpage :
4714
Abstract :
We consider the problem of maximizing underwater acoustic data transmission, by adaptively positioning a mobile relay. This is a classic exploration vs. exploitation scenario well-described by a multi-armed bandit formulation, which in its canonical form is optimally solved by the Gittins index rule. For an ocean vehicle traveling between distant waypoints, however, switching costs are significant, and the MAB with switching costs has no optimal index policy. To address this we have developed a strong adaptation of the Gittins index rule that employs limited-horizon enumeration. We describe autonomous shallow-water field experiments conducted in the Charles River (Boston, MA) with unmanned vehicles and acoustic modems, and compare the performance of different algorithms. Our switching-costs-aware MAB heuristic offers both superior real-time performance in decision-making and efficient learning of the unknown field.
Keywords :
data communication; decision making; modems; telecommunication switching; underwater acoustic communication; Boston; Gittins index rule; acoustic modems; acoustic relay positioning; bandit methods performance; decision making; mobile; multiarmed bandit formulation; ocean vehicle; switching costs; switching-costs-aware MAB heuristic; underwater acoustic data transmission; unmanned vehicles; Acoustics; Indexes; Modems; Relays; Rivers; Switches; Vehicles; (Under)water vehicles; Autonomous systems; Optimization algorithms;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
American Control Conference (ACC), 2014
Conference_Location :
Portland, OR
ISSN :
0743-1619
Print_ISBN :
978-1-4799-3272-6
Type :
conf
DOI :
10.1109/ACC.2014.6859385
Filename :
6859385
Link To Document :
بازگشت