Presentation Title

Intelligently Searching and Mapping Shipwrecks using Autonomous Underwater Vehicles

Start Date

November 2016

End Date

November 2016

Location

HUB 379

Type of Presentation

Oral Talk

Abstract

The goal of this project is to aid marine archaeologists by detecting and mapping archaeological sites using Autonomous Underwater Vehicles (AUVs). There are two main aspects: shipwreck search and shipwreck mapping. Shipwreck search uses image processing techniques on side-scan sonar data to identify archaeological sites. The shipwreck mapping component is using motion planning algorithms to plan optimal paths around an archaeological site to collect camera data. The camera data is used to create 3D reconstructions of the archaeological sites.

The motion planning algorithm is a variation of the rapidly-exploring random tree (RRT) algorithm. Several modifications to the algorithm have been made and simulation shows up to 152% improvement in information gain. Furthermore, this algorithm has been successfully tested on two archaeological sites off the coast of Malta. The team was able to create 3D reconstructions of the Manoel Island X-Lighter shipwreck and the Bristol Beaufighter plane wreck.

This document is currently not available here.

Share

COinS
 
Nov 12th, 3:15 PM Nov 12th, 3:30 PM

Intelligently Searching and Mapping Shipwrecks using Autonomous Underwater Vehicles

HUB 379

The goal of this project is to aid marine archaeologists by detecting and mapping archaeological sites using Autonomous Underwater Vehicles (AUVs). There are two main aspects: shipwreck search and shipwreck mapping. Shipwreck search uses image processing techniques on side-scan sonar data to identify archaeological sites. The shipwreck mapping component is using motion planning algorithms to plan optimal paths around an archaeological site to collect camera data. The camera data is used to create 3D reconstructions of the archaeological sites.

The motion planning algorithm is a variation of the rapidly-exploring random tree (RRT) algorithm. Several modifications to the algorithm have been made and simulation shows up to 152% improvement in information gain. Furthermore, this algorithm has been successfully tested on two archaeological sites off the coast of Malta. The team was able to create 3D reconstructions of the Manoel Island X-Lighter shipwreck and the Bristol Beaufighter plane wreck.