This research focuses on bridging the scale difference in data collection from small defects of large civil infrastructures. The idea is to use the domain knowledge and artificial intelligence (AI) to identify indications of defects in low resolution scans and prioritizes high-resolution robotic scans accordingly.
Our first paper in this area introduces a multi-scale robotic approach for measuring the width, length, shape, and profile of hairline cracks in concrete structures. The approach uses a convolutional neural network to identify potential surface cracks, and then robotically navigates a high-resolution laser scanner to measure the detailed shape of the detected cracks. Finally, 3D point cloud registration techniques fuse the laser scans with LiDAR-based scan of the surrounding environment. The proposed method is validated with computer simulations and physical experiments on a concrete specimen. The results are compared against the state-of-the-art, vision-based methods as well as readings of a transparent crack width ruler. The comparison demonstrates the superiority and effectiveness of the proposed multi-scale robotic approach in measuring hairline cracks providing vital data for assessing the conditions of civil infrastructures.
For more details on our research, we encourage readers to access our paper in the “Automation in Construction” journal:
Ali Ghadimzadeh Alamdari and Arvin Ebrahimkhanlou “A Multi-Scale Robotic Approach for Precise Crack Measurement in Concrete Structures”, Automation in Construction 158, No. 105215 (Feb 2024) https://doi.org/10.1016/j.autcon.2023.105215
Reinforced concrete structures are no strangers to the phenomenon of cracking. Natural aging, environmental factors, and various loading conditions contribute to the formation and growth of these cracks. While some are harmless, others can be harbingers of compromised structural integrity or even catastrophic failure. The analogy is striking – much like a patient’s medical symptoms, these cracks demand early detection and accurate measurement to ensure the structural health of buildings, bridges, and other infrastructures.
Our study presents a robotic approach to this challenge, employing a robotic arm integrated with a laser scanner. This robotic system offers an unbiased, consistent, and thorough examination that may be hard to reach in manual inspections. Unlike human inspectors, these robots can tirelessly scan structures, employing various non-destructive testing techniques to detect even the smallest defects.
However, this method is not without its challenges. High-resolution data collection from vast structures can be overwhelming in terms of computational resources, time, and budget. Our research addresses this by prioritizing high-resolution data collection from areas exhibiting defects and maintaining a lower resolution, yet sufficiently accurate, digital twin of the structure. This multi-scale strategy is not only time-efficient but also ensures that no critical data is overlooked.
News coverage:
1) Newswise, “Drexel Researchers Propose AI-Guided System for Robotic Inspection of Buildings, Roads and Bridges,” January 31, 2024, https://www.newswise.com/articles/drexel-researchers-propose-ai-guided-system-for-robotic-inspection-of-buildings-roads-and-bridges
2) LiDAR News, “AI and Lidar for Robotic Infrastructure Inspection,” February 02, 2024 https://blog.lidarnews.com/ai-and-lidar-for-robotic-infrastructure-inspection/
3) COSMOS “Robots could be inspecting our buildings one day,” February 05, 2024 https://cosmosmagazine.com/science/engineering/you-might-have-missed-improving-bowel-cancer-treatment-robot-inspectors-the-first-continents-and-fledgling-planets/
4) Construction Dive “Drexel researchers create AI system to spot cracks in infrastructure,” Matthew Thibault https://www.constructiondive.com/news/drexel-robot-inspections-infrastructure-concrete-cracks/706820/ February 07, 2024