No Cover Image

Conference Paper/Proceeding/Abstract 101 views 1 download

Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics

Michael Watson, Hans Ren, Farshad Arvin, Junyan Hu

Towards Autonomous Robotic Systems

Swansea University Author: Hans Ren

  • Michael_TAROS.pdf

    PDF | Accepted Manuscript

    Author accepted manuscript document released under the terms of a Creative Commons CC-BY licence using the Swansea University Research Publications Policy (rights retention).

    Download (627KB)

Abstract

Coverage Path Planning (CPP) is an effective approach to let intelligent robots cover an area by finding feasible paths through the environment. In this paper, we focus on using reinforcement learning to learn about a given environment and find the most efficient path that explores all target points...

Full description

Published in: Towards Autonomous Robotic Systems
Published: 2024
URI: https://cronfa.swan.ac.uk/Record/cronfa66908
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2024-06-28T18:29:17Z
last_indexed 2024-06-28T18:29:17Z
id cronfa66908
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>66908</id><entry>2024-06-28</entry><title>Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics</title><swanseaauthors><author><sid>9e043b899a2b786672a28ed4f864ffcc</sid><firstname>Hans</firstname><surname>Ren</surname><name>Hans Ren</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-06-28</date><deptcode>MACS</deptcode><abstract>Coverage Path Planning (CPP) is an effective approach to let intelligent robots cover an area by finding feasible paths through the environment. In this paper, we focus on using reinforcement learning to learn about a given environment and find the most efficient path that explores all target points. To overcome the limitations caused by stan- dard Q-learning based CPP that often fall into a local optimum and may be in-efficient in large-scale environments, two methods of improve- ment are considered, i.e., the use of a robot swarm working towards the same goal and the augmenting of the Q-learning algorithm to include a predator-prey based reward system. Existing predator-prey based reward systems provide rewards the further away an agent is from its predator, the paper adapts this concept to work within a robot swarm by simulat- ing each agent of the swarm as both predator and prey. Simulation case studies and comparisons with the standard Q-learning show that the proposed method has a superior coverage performance in complicated environments.</abstract><type>Conference Paper/Proceeding/Abstract</type><journal>Towards Autonomous Robotic Systems</journal><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic/><keywords/><publishedDay>28</publishedDay><publishedMonth>6</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-06-28</publishedDate><doi/><url/><notes/><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm/><funders/><projectreference/><lastEdited>2024-08-08T13:59:13.1037536</lastEdited><Created>2024-06-28T19:24:58.2731592</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Michael</firstname><surname>Watson</surname><order>1</order></author><author><firstname>Hans</firstname><surname>Ren</surname><order>2</order></author><author><firstname>Farshad</firstname><surname>Arvin</surname><order>3</order></author><author><firstname>Junyan</firstname><surname>Hu</surname><order>4</order></author></authors><documents><document><filename>66908__30779__04553fc7c7c942afa64a946d8eeb66d8.pdf</filename><originalFilename>Michael_TAROS.pdf</originalFilename><uploaded>2024-06-28T19:28:46.8381822</uploaded><type>Output</type><contentLength>642046</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2024-07-28T00:00:00.0000000</embargoDate><documentNotes>Author accepted manuscript document released under the terms of a Creative Commons CC-BY licence using the Swansea University Research Publications Policy (rights retention).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/deed.en</licence></document></documents><OutputDurs/></rfc1807>
spelling v2 66908 2024-06-28 Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics 9e043b899a2b786672a28ed4f864ffcc Hans Ren Hans Ren true false 2024-06-28 MACS Coverage Path Planning (CPP) is an effective approach to let intelligent robots cover an area by finding feasible paths through the environment. In this paper, we focus on using reinforcement learning to learn about a given environment and find the most efficient path that explores all target points. To overcome the limitations caused by stan- dard Q-learning based CPP that often fall into a local optimum and may be in-efficient in large-scale environments, two methods of improve- ment are considered, i.e., the use of a robot swarm working towards the same goal and the augmenting of the Q-learning algorithm to include a predator-prey based reward system. Existing predator-prey based reward systems provide rewards the further away an agent is from its predator, the paper adapts this concept to work within a robot swarm by simulat- ing each agent of the swarm as both predator and prey. Simulation case studies and comparisons with the standard Q-learning show that the proposed method has a superior coverage performance in complicated environments. Conference Paper/Proceeding/Abstract Towards Autonomous Robotic Systems 28 6 2024 2024-06-28 COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University 2024-08-08T13:59:13.1037536 2024-06-28T19:24:58.2731592 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Michael Watson 1 Hans Ren 2 Farshad Arvin 3 Junyan Hu 4 66908__30779__04553fc7c7c942afa64a946d8eeb66d8.pdf Michael_TAROS.pdf 2024-06-28T19:28:46.8381822 Output 642046 application/pdf Accepted Manuscript true 2024-07-28T00:00:00.0000000 Author accepted manuscript document released under the terms of a Creative Commons CC-BY licence using the Swansea University Research Publications Policy (rights retention). true eng https://creativecommons.org/licenses/by/4.0/deed.en
title Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics
spellingShingle Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics
Hans Ren
title_short Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics
title_full Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics
title_fullStr Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics
title_full_unstemmed Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics
title_sort Predator-Prey Q-Learning Based Collaborative Coverage Path Planning for Swarm Robotics
author_id_str_mv 9e043b899a2b786672a28ed4f864ffcc
author_id_fullname_str_mv 9e043b899a2b786672a28ed4f864ffcc_***_Hans Ren
author Hans Ren
author2 Michael Watson
Hans Ren
Farshad Arvin
Junyan Hu
format Conference Paper/Proceeding/Abstract
container_title Towards Autonomous Robotic Systems
publishDate 2024
institution Swansea University
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Coverage Path Planning (CPP) is an effective approach to let intelligent robots cover an area by finding feasible paths through the environment. In this paper, we focus on using reinforcement learning to learn about a given environment and find the most efficient path that explores all target points. To overcome the limitations caused by stan- dard Q-learning based CPP that often fall into a local optimum and may be in-efficient in large-scale environments, two methods of improve- ment are considered, i.e., the use of a robot swarm working towards the same goal and the augmenting of the Q-learning algorithm to include a predator-prey based reward system. Existing predator-prey based reward systems provide rewards the further away an agent is from its predator, the paper adapts this concept to work within a robot swarm by simulat- ing each agent of the swarm as both predator and prey. Simulation case studies and comparisons with the standard Q-learning show that the proposed method has a superior coverage performance in complicated environments.
published_date 2024-06-28T13:59:12Z
_version_ 1806824323971809280
score 11.021648

Similar Items