Presentation + Paper
9 May 2024 Multi-scale robotic scanning of surface cracks in concrete structures
Ali Ghadimzadeh Alamdari, Arvin Ebrahimkhanlou
Author Affiliations +
Abstract
Defects such as surface cracks have considerably smaller scales in civil infrastructure, despite the grand scope of the structures themselves. This article introduces a robotic technique designed to bridge this scale gap by measuring the dimensions and characteristics of small-scale surface cracks in concrete structures at multiple scales. Using a convolutional neural network, our system initially detects potential surface cracks - regions of interests (ROI). Once identified, a high-definition laser scanner, steered by a robotic arm, scans the geometry of these ROIs. The detailed laser scan data is subsequently integrated with surrounding large-scale environmental scans obtained via LiDAR, utilizing 3D point cloud alignment methods. We validate the proposed solution through both computer simulations using the robotic operation system (ROS) as well as testing on a physical concrete specimen. Our method offers an unprecedented resolution of 0.004 mm for crack width measurement and has been successfully tested on real-world cracks with a width of 0.17 mm. A comparative analysis with existing vision-based solutions and traditional crack-width measuring instruments confirms the superior accuracy and efficiency of our multi-scale robotic approach in deriving crucial metrics necessary for assessing the structural health of civil infrastructure.
Conference Presentation
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Ali Ghadimzadeh Alamdari and Arvin Ebrahimkhanlou "Multi-scale robotic scanning of surface cracks in concrete structures", Proc. SPIE 12950, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XVIII, 129500M (9 May 2024); https://doi.org/10.1117/12.3010480
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Robotics

Laser scanners

Point clouds

Inspection

Cameras

LIDAR

Data modeling

Back to Top