Back to search

Georgia Institute of Technology

Robotic crack detection and radiation simulation in a virtual nuclear spent fuel pool

Abstract

dc:description.abstract

The use of robotics and Machine Learning (ML) in Nuclear Power Plants (NPPs) continues to grow, but these technologies must be thoroughly evaluated before they can be used in safety-critical settings. Inspection robots intended for radiological or restricted areas need to be assessed not only for basic mobility, but also for sensing, perception, and overall suitability for realistic plant tasks. Conducting this kind of testing in an operating facility is expensive, difficult to repeat, and often constrained by plant operations. For that reason, a high-fidelity virtual environment is valuable as a way to evaluate robotic inspection tasks before deployment. The goal of this thesis is to enable repeatable evaluation of robot sensing, perception, and inspection performance for spent fuel pool inspection without disrupting plant operations. To achieve this, the thesis develops a radiation-aware simulation environment for spent fuel pool inspection and integrates a real-time crack classification method for vision-based inspections. A key novelty of this work is the creation of a virtual spent fuel pool environment for underwater robotic inspection testing that integrates robot simulation, crack inspection, and radiation-aware operation. This framework integrates a Unreal Engine 5 (UE5) virtual environment, an underwater Remote Operated Vehicle (ROV) simulated through UNav-Sim, a hybrid radiation simulation derived from OpenMC, and a Convolutional Neural Network (CNN)-based crack classification pipeline. Together, these elements provide a virtual testbed for developing, rehearsing, and evaluating inspection tasks without exposing personnel or disrupting plant operations.

Degree

thesis:*
Name thesis:degree_name
Nuclear Engineering, MS
Department thesis:degree_department
Mechanical Engineering
Grantor
Georgia Institute of Technology
Year dc:date.issued
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Banks, Spencer
Advisor dc:contributor.advisor
  • Zhang, Fan
Committee members dc:contributor.committeemember
  • Biegalski, Steven
  • Che, Yifeng

Rights

Language dc:language.iso
English

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/81625

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Banks, Spencer. Robotic crack detection and radiation simulation in a virtual nuclear spent fuel pool. Georgia Institute of Technology, 2026. https://hdl.handle.net/1853/81625