A graphics-based digital twin framework for computer vision-based post-earthquake structural inspection and evaluation using unmanned aerial vehicles

Shuo Wang, Casey Rodgers, Guanghao Zhai, Thomas Ngare Matiki, Brian Welsh, Amirali Najafi, Jingjing Wang, Yasutaka Narazaki, Vedhus Hoskere, Billie F. Spencer

Research output: Contribution to journalArticlepeer-review

Abstract

Rapid structural inspections and evaluations are critical after earthquakes. Computer vision-based methods have attracted the interest of researchers for their potential to be rapid, safe, and objective. To provide an end-to-end solution for computer vision-based post-earthquake inspection and evaluation of a specific as-built structure, the concepts of physics-based graphics model (PBGM) and digital twin (DT) are combined to develop a graphics-based digital twin (GBDT) framework. The GBDT framework comprises a finite element (FE) model and a computer graphics (CG) model whose state is informed by the FE analysis, representing the state of the structure before and after an earthquake. The CG model is first created making use of the FE model and the photographic survey of the structure, yielding the virtual counterpart of the as-built structure quickly and accurately. Then damage modelling approaches are proposed to predict the location and extent of structural and nonstructural damage under seismic loading, from which photographic representation of the predicted damage is realized in the CG model. The effectiveness of the GBDT framework is demonstrated using a five-story reinforced concrete benchmark building through the design and assessment of various UAV (Unmanned Aerial Vehicle) inspection trajectories for post-earthquake scenarios. The results demonstrate that the proposed GBDT framework has significant potential to enable rapid structural inspection and evaluation, ultimately leading to more efficient allocation of scarce resources in a post-earthquake setting.

Original languageEnglish (US)
Article number100003
JournalJournal of Infrastructure Intelligence and Resilience
Volume1
Issue number1
DOIs
StatePublished - Sep 2022

Keywords

  • Computer vision
  • Digital twin
  • Earthquake engineering
  • Graphics-based digital twin
  • Physics-based graphics model
  • Post-earthquake assessment

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'A graphics-based digital twin framework for computer vision-based post-earthquake structural inspection and evaluation using unmanned aerial vehicles'. Together they form a unique fingerprint.

Cite this