Representative Research Projects
Virtual Robotic Bridge Inspection
As robots become a more accepted tool for infrastructure inspection, it becomes increasingly important to consider how a human will interact with the captured information. This is fundamentally a data representation problem. We are studying how to convert robotic inspection information into virtual computer “worlds” that can be explored by a human inspector. To accomplish this, we utilize techniques adapted from computer vision (such as Structure From Motion) and virtual reality equipment such as the Oculus Rift.
Structural Evaluation Through Computer Vision
After a disaster, such as an earthquake, inspectors are tasked with assessing the integrity of any and all affected structures. Depending on the scale of the disaster, the number of required inspections can range into the thousands, and so rapid assessments are vital. The goal of this project is to develop a method of assessing the integrity of civil structures using a combination of digital image analysis and artificial intelligence. Fully realized, the proposed technology will enable post-disaster inspectors to rapidly and accurately estimate structural damage using only a digital camera and portable computer.
Extracting Finite Element Models from Images
Related to our efforts in structural evaluation, we are studying ways of automatically extracting finite element models from collections of images. Using a combination of Structure from Motion (SfM) algorithms and 3D computer vision, we are working on ways to automatically recognize and assemble structural components, followed by FE model generation and updating based on damage detected in the image collections. The proposed methodology has applications beyond post-disaster condition assessment, from routine inspection to infrastructure management .
3D Printing for Engineering Education
Hands-on laboratory activities and demonstrations are an essential component of educational programs in every field of engineering. However, financial and logistical practicalities often inhibit students from being able to actualize and experiment on their own designs. We are studying how to use 3D printing to illustrate a wide variety of engineering concepts, and to improve learning outcomes for civil engineering students in particular. Shown is LRG student Jeff Bynum with his 3D printed version of the AISC steel sculpture.