ORCID

0009-0009-0001-4406

Date of Award

2025

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Mechanical Engineering

First Advisor

Dryver Huston

Abstract

Maintenance and inspection of small culverts are vital components of stormwater and flood-resilient infrastructure. Small culverts being defined as less than 1 m in diameter prevent easy and safe human access. This research explores methods for small culvert inspection using low-cost, easy-to-build, semi-autonomous swarm robots. Culverts, being radio absorptive and GPS denied, present significant telemetry and navigation challenges. The swarm of autonomous follower robots enable inspection of long and complex culverts, utilizing each robot in the swarm as a relay node in a network to transmit video and control signals between the user and the lead robot. Two leader tracking methods are studied. The first utilizes RGB-D video for intelligent leader and obstacle detection and following. The second utilizes only RGB or grayscale video for leader detection and following, at the cost of obstacle detection but benefiting from significantly less expensive hardware: about 1\% the cost of typical culvert inspection robots. Another primary development is of an inexpensive universal dual motor control board and simple methods to tune its control algorithm to any differential drive electric vehicle. The entire system is built on a scalable and modular platform capable of being used in search and rescue, structural health monitoring, and radar surveying using additional sensors.

Language

en

Number of Pages

142 p.

Available for download on Thursday, April 22, 2027

Share

COinS