Strathprints logo
Strathprints Home | Open Access | Browse | Search | User area | Copyright | Help | Library Home | SUPrimo

An emergent wall following behaviour to escape local minima for swarms of agents

Abdel Wahid, Mohamed Hussien Mabrouk and McInnes, C.R. (2008) An emergent wall following behaviour to escape local minima for swarms of agents. International Journal of Computer Science, 35 (4). IJCS-35. ISSN 1819-656X

[img]
Preview
PDF (strathprints008109.pdf)
strathprints008109.pdf

Download (653kB) | Preview

Abstract

Natural examples of emergent behaviour, in groups due to interactions among the group's individuals, are numerous. Our aim, in this paper, is to use complex emergent behaviour among agents that interact via pair-wise attractive and repulsive potentials, to solve the local minima problem in the artificial potential based navigation method. We present a modified potential field based path planning algorithm, which uses agent internal states and swarm emergent behaviour to enhance group performance. The algorithm is used successfully to solve a reactive path-planning problem that cannot be solved using conventional static potential fields due to local minima formation. Simulation results demonstrate the ability of a swarm of agents to perform problem solving using the dynamic internal states of the agents along with emergent behaviour of the entire group.

Item type: Article
ID code: 8109
Keywords: agent internal states, local minima escape, swarm emergent behaviour, wall following, navigation systems, Mechanical engineering and machinery, Motor vehicles. Aeronautics. Astronautics, Aerospace Engineering, Control and Systems Engineering, Computational Mechanics
Subjects: Technology > Mechanical engineering and machinery
Technology > Motor vehicles. Aeronautics. Astronautics
Department: Faculty of Engineering > Mechanical and Aerospace Engineering
Depositing user: Ms Katrina May
Date Deposited: 22 Jun 2009 11:44
Last modified: 21 May 2015 10:41
Related URLs:
URI: http://strathprints.strath.ac.uk/id/eprint/8109

Actions (login required)

View Item View Item