Title: Design of genetic algorithms for topology control of unmanned vehicles

Authors: Cem Safak Sahin, Elkin Urrea, M. Umit Uyar, Michael Conner, Giorgio Bertoli, Christian Pizzo

Addresses: Department of Electrical Engineering, 138th Street and Convent Avenue, City College of the City University of New York, New York, NY 10031, USA. ' Department of Electrical Engineering, 138th Street and Convent Avenue, City College of the City University of New York, New York, NY 10031, USA. ' Department of Electrical Engineering, 138th Street and Convent Avenue, City College of the City University of New York, New York, NY 10031, USA. ' Department of Electrical Engineering, 138th Street and Convent Avenue, City College of the City University of New York, New York, NY 10031, USA. ' US Army Communications-Electronic RD&E Center, Offensive Information Operations (OIO) Branch, Ft. Monmouth, NJ 07703, USA. ' US Army Communications-Electronic RD&E Center, Offensive Information Operations (OIO) Branch, Ft. Monmouth, NJ 07703, USA

Abstract: We present genetic algorithms (GAs) as a decentralised topology control mechanism distributed among active running software agents to achieve a uniform spread of terrestrial unmanned vehicles (UVs) over an unknown geographical area. This problem becomes more challenging under the harsh and bandwidth limited conditions of military applications. Using only local neighbour information, a GA guides each UV to select a |fitter| speed and direction among exponentially large number of choices, converging towards a uniform node distribution. In an observed occurrence of a threat situation during a mission where UVs are to spread uniformly over an unknown terrain, if the number of UVs change with time (e.g., losing assets due to hostile forces), the remaining units should reposition themselves to compensate the loss in area coverage. Our simulation software results show that GAs can be an effective tool for providing a robust solution for topology control of UVs in military applications.

Keywords: genetic algorithms; GAs; artificial intelligence; mobile ad hoc networks; MANETs; bio-inspired algorithms; unmanned vehicles; UVs; topology control; mobile networks; decentralised control; military applications; simulation; unknown terrain.

DOI: 10.1504/IJADS.2010.036100

International Journal of Applied Decision Sciences, 2010 Vol.3 No.3, pp.221 - 238

Published online: 19 Oct 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article