Automated Close-Quarter, High-Resolution Inspection and 3D Reconstruction of Unknown Infrastructure

Ben Bartlett, Petar Trslic, Matheus Santos, Manduhu Manduhu, James Riordan, Gerard Dooly

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper presents a new automated workflow for 3D reconstruction of unknown infrastructure utilising UAS technology, developed within the EU funded RAPID project. A multi stage approach is used to produce highly accurate reconstructions of infrastructure, specifically unknown infrastructure. Using this approach, the drone operator is not required to perform extensive pre-flight surveying and mission planning, adding only a few high level inputs to the overall workflow. This benefits inspection workflows where the pilot is not onsite and cannot make visual observations on safety. Furthermore, this results in uniform high resolution reconstructions and a higher level of consistent and simple task loading for operator. Real-world environment tests consisting of using one of the most popular commercial UAS platforms to perform autonomous surveys was completed, with results to demonstrate the feasibility of the developed system.

Original languageEnglish
Title of host publicationOCEANS 2023 - Limerick, OCEANS Limerick 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350332261
DOIs
Publication statusPublished - 2023
Event2023 OCEANS Limerick, OCEANS Limerick 2023 - Limerick, Ireland
Duration: 5 Jun 20238 Jun 2023

Publication series

NameOCEANS 2023 - Limerick, OCEANS Limerick 2023

Conference

Conference2023 OCEANS Limerick, OCEANS Limerick 2023
Country/TerritoryIreland
CityLimerick
Period5/06/238/06/23

Keywords

  • 3D Reconstruction
  • UAS
  • UAV
  • autonomous workflow
  • uniform GSD
  • unknown environments

Fingerprint

Dive into the research topics of 'Automated Close-Quarter, High-Resolution Inspection and 3D Reconstruction of Unknown Infrastructure'. Together they form a unique fingerprint.

Cite this