Mostrar el registro sencillo del ítem

dc.contributor.authorRodrigo, Daniel Vicente
dc.contributor.authorSierra Garcia, Jesús Enrique 
dc.contributor.authorSantos, Matilde
dc.date.accessioned2023-03-02T11:15:32Z
dc.date.available2023-03-02T11:15:32Z
dc.date.issued2023-01
dc.identifier.issn0965-9978
dc.identifier.urihttp://hdl.handle.net/10259/7483
dc.description.abstractThe COVID-19 pandemic made robot manufacturers explore the idea of combining mobile robotics with UV-C light to automate the disinfection processes. But performing this process in an optimum way introduces some challenges: on the one hand, it is necessary to guarantee that all surfaces receive the radiation level to ensure the disinfection; at the same time, it is necessary to minimize the radiation dose to avoid the damage of the environment. In this work, both challenges are addressed with the design of a complete coverage path planning (CCPP) algorithm. To do it, a novel architecture that combines the glasius bio-inspired neural network (GBNN), a motion strategy, an UV-C estimator, a speed controller, and a pure pursuit controller have been designed. One of the main issues in CCPP is the deadlocks. In this application they may cause a loss of the operation, lack of regularity and high peaks in the radiation dose map, and in the worst case, they can make the robot to get stuck and not complete the disinfection process. To tackle this problem, in this work we propose a preventive deadlock processing algorithm (PDPA) and an escape route generator algorithm (ERGA). Simulation results show how the application of PDPA and the ERGA allow to complete complex maps in an efficient way where the application of GBNN is not enough. Indeed, a 58% more of covered surface is observed. Furthermore, two different motion strategies have been compared: boustrophedon and spiral motion, to check its influence on the performance of the robot navigation.en
dc.format.mimetypeapplication/pdf
dc.language.isoenges
dc.publisherElsevieren
dc.relation.ispartofAdvances in Engineering Software. 2023, V. 17, 103330en
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 Internacional*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectComplete coverage path planningen
dc.subjectMobile roboten
dc.subjectUV-Cen
dc.subjectDeadlocksen
dc.subjectEscape routesen
dc.subject.otherElectrotecniaes
dc.subject.otherElectrical engineeringen
dc.subject.otherIngeniería mecánicaes
dc.subject.otherMechanical engineeringen
dc.titleGlasius bio-inspired neural networks based UV-C disinfection path planning improved by preventive deadlock processing algorithmen
dc.typeinfo:eu-repo/semantics/articlees
dc.rights.accessRightsinfo:eu-repo/semantics/openAccesses
dc.relation.publisherversionhttps://doi.org/10.1016/j.advengsoft.2022.103330es
dc.identifier.doi10.1016/j.advengsoft.2022.103330
dc.journal.titleAdvances in Engineering Softwareen
dc.volume.number175es
dc.page.initial103330es
dc.type.hasVersioninfo:eu-repo/semantics/publishedVersiones


Ficheros en este ítem

Thumbnail

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem