Candan, F., Peng, Y. and Mihaylova, L. orcid.org/0000-0001-5856-2223 (2021) A comparison of obstacle dependant Gaussian and hybrid potential field methods for collision avoidance in multi-agent systems. In: Jamil, A. and Hameed, A.A., (eds.) Proceedings of the 1st International Conference on Computing and Machine Intelligence (ICMI 2021). 1st International Conference on Computing and Machine Intelligence (ICMI-21), 19-20 Feb 2021, Virtual conference. Istanbul Sabahattin Zaim University Yayınları , pp. 240-245. ISBN 978-605-06675-7-8
Abstract
In this paper collision avoidance methods - the velocity obstacle, obstacle dependant Gaussian potential field and hybrid potential field methods, are compared and tested on multiagent systems. Extensive evaluation is presented with a number of case studies with a different number of agents, with static and dynamic and obstacles. The advantages and disadvantages of each method are discussed. The optimisation of the static and dynamic coefficients of the hybrid potential field method is performed via a genetic algorithm. The results from the tests are from 1000 independent runs and show that the hybrid potential field method can avoid reliably collisions in multi-agent systems.
Metadata
Item Type: | Proceedings Paper |
---|---|
Authors/Creators: |
|
Editors: |
|
Copyright, Publisher and Additional Information: | © 2021. |
Keywords: | multi-agent systems; unmanned air vehicles; collision avoidance; potential field |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Automatic Control and Systems Engineering (Sheffield) |
Depositing User: | Symplectic Sheffield |
Date Deposited: | 12 Feb 2021 10:52 |
Last Modified: | 16 Oct 2023 09:20 |
Published Version: | https://icmi.aiplustech.org/icmi21_home.html |
Status: | Published |
Publisher: | Istanbul Sabahattin Zaim University Yayınları |
Refereed: | Yes |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:171034 |
Download
Filename: ICMI2021 Final version.pdf
