UWROV

I am one of two software leads on the Underwater Remotely Operated Vehicles Team at the University of Washington. This team focuses on underwater robots and is a world-class team in terms of robot performance and documentation.
My role as a software developer on this team has largely been in the realm of computer vision, both in the fields of Structure from Motion (underwater position localization) and Photogrammetry (creating 3D models of the sea floor and structures on it). Below is an example of some of the 3D reconstruction work I've been doing. This image features a pipe structure used to mimick coral, captured from a camera mounted on the bottom of the ROV.
At the 2025 world championships, we won 3 awards -- we were the Engineering Presentation World Champions, Technical Documentation World Champions, and Sharkpedo Legacy Innovation Award winners.
In the summer of 2025, we participated in the MATE ROV / NOAA Ocean Exploration Video Challenge, an AI and CV challenge. I lead the development on this, and we ended up winning 1st place, which won us a sponsored trip to the Underwater Intervention Conference to present our work.