Design of a Robotic Inspection Platform for Structural Health Monitoring

Authors

Jason R. Soto, Florida International UniversityFollow

Document Type

Thesis

Degree

Master of Science (MS)

Major/Program

Mechanical Engineering

First Advisor's Name

Dwayne McDaniel

First Advisor's Committee Title

Committee chair

Second Advisor's Name

Leonel Lagos

Second Advisor's Committee Title

committee member

Third Advisor's Name

Ibrahim Tansel

Third Advisor's Committee Title

committee member

Keywords

robot, autonomous, ros, slam, lidar, point, cloud, monitoring, infrastructure, mapping

Date of Defense

6-29-2020

Abstract

Actively monitoring infrastructure is key to detecting and correcting problems before they become costly. The vast scale of modern infrastructure poses a challenge to monitoring due to insufficient personnel. Certain structures, such as refineries, pose additional challenges and can be expensive, time-consuming, and hazardous to inspect.

This thesis outlines the development of an autonomous robot for structural-health-monitoring. The robot is capable of operating autonomously in level indoor environments and can be controlled manually to traverse difficult terrain. Both visual and lidar SLAM, along with a procedural-mapping technique, allow the robot to capture colored-point-clouds.

The robot is successfully able to automate the point cloud collection of straightforward environments such as hallways and empty rooms. While it performs well in these situations, its accuracy suffers in complex environments with variable lighting. More work is needed to create a robust system, but the potential time savings and upgrades make the concept promising.

Identifier

FIDC009150

Recommended Citation

Soto, Jason R., "Design of a Robotic Inspection Platform for Structural Health Monitoring" (2020). FIU Electronic Theses and Dissertations. 4483.
https://digitalcommons.fiu.edu/etd/4483