Close menu


Sunderland Repository records the research produced by the University of Sunderland including practice-based research and theses.

A novel potential field model for perimeter and agent density control in multiagent swarms

Eliot, N, Kendall, D, Brockway, M, Oman, P and Bouridane, A (2023) A novel potential field model for perimeter and agent density control in multiagent swarms. Expert Systems With Applications, 227. ISSN 0957-4174

Item Type: Article


parameters for the computation of control vectors. This restriction often limits the structures that can evolve,
since agents are unable to modify their behaviour based on their structural role. This paper proposes an
enhanced model that uses the perimeter status of agents in selecting control parameters. This allows a wider
variety of emergent behaviours, many of which result in much improved swarm structures. The model is
based upon equivalence classes of agent pairs, defined by their perimeter status. Array-valued parameters are
introduced to allow each equivalence class to be given its own parameter values. The model also introduces a
new control vector to ‘flatten’ reflex angles between neighbouring agents on the swarm perimeter, often leading
to significantly improved swarm structure. Extensive experiments have been conducted that demonstrate how
the new model causes a variety of useful behaviours to emerge from random swarm deployments. The results
show that several important behaviours, such as shape control, void removal, perimeter packing and expansion,
and perimeter rotation, can be produced without the need for explicit inter-agent communication. The
approach is applicable to a variety of applications, including reconnaissance, area-coverage, and containment.

PDF (A novel potential field model for perimeter and agent density control in multiagent swarms)
1-s2.0-S0957417423006851-main.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview

More Information

Depositing User: Neil Eliot


Item ID: 17208
Identification Number:
ISSN: 0957-4174
Official URL:

Users with ORCIDS

ORCID for N Eliot: ORCID iD
ORCID for D Kendall: ORCID iD
ORCID for M Brockway: ORCID iD
ORCID for P Oman: ORCID iD
ORCID for A Bouridane: ORCID iD

Catalogue record

Date Deposited: 15 Jan 2024 09:36
Last Modified: 15 Jan 2024 09:45


Author: N Eliot ORCID iD
Author: D Kendall ORCID iD
Author: M Brockway ORCID iD
Author: P Oman ORCID iD
Author: A Bouridane ORCID iD

University Divisions

Faculty of Technology > School of Computer Science


Computing > Artificial Intelligence

Actions (login required)

View Item (Repository Staff Only) View Item (Repository Staff Only)