Control of Agent Swarms using Generalized Centroidal Cyclic Pursuit Laws

Sinha, A and Ghose, D (2007) Control of Agent Swarms using Generalized Centroidal Cyclic Pursuit Laws. In: Workshop on Analytics for Noisy Unstructured Text Data held in Conjunction with the 20th International Joint Conference on Artificial Intelligence, JAN 06-12, 2007 , Hyderabad, INDIA, pp. 1525-1530.

PDF Control.pdf - Published Version Restricted to Registered users only Download (239kB) | Request a copy

Abstract

One of the major tasks in swarm intelligence is to design decentralized but homogenoeus strategies to enable controlling the behaviour of swarms of agents. It has been shown in the literature that the point of convergence and motion of a swarm of autonomous mobile agents can be controlled by using cyclic pursuit laws. In cyclic pursuit, there exists a predefined cyclic connection between agents and each agent pursues the next agent in the cycle. In this paper we generalize this idea to a case where an agent pursues a point which is the weighted average of the positions of the remaining agents. This point correspond to a particular pursuit sequence. Using this concept of centroidal cyclic pursuit, the behavior of the agents is analyzed such that, by suitably selecting the agents' gain, the rendezvous point of the agents can be controlled, directed linear motion of the agents can be achieved, and the trajectories of the agents can be changed by switching between the pursuit sequences keeping some of the behaviors of the agents invariant. Simulation experiments are given to support the analytical proofs.

Item Type:	Conference Paper
Publisher:	Ijcai-Int Conf Artif Intell
Additional Information:	Copyright of this article belongs to Ijcai-Int Conf Artif Intell.
Department/Centre:	Division of Mechanical Sciences > Aerospace Engineering(Formerly Aeronautical Engineering)
Date Deposited:	12 Jan 2011 11:04
Last Modified:	02 Nov 2011 06:19
URI:	http://eprints.iisc.ac.in/id/eprint/34915

Actions (login required)

View Item