Image courtesy of iCORE Lab, Louisiana State University. Figure C shows initial 3D reconstructions obtained from data collected by the ROV.įigure C: Preliminary 3D reconstruction of Monohansett obtained with data from an ROV. With data collected by the ROV, we are able to create large-scale 3D reconstructions of shipwreck sites. Over the course of the expedition, we had a total of 13 ROV deployments with three different ROV platforms. We also conducted preliminary deployments of a remotely operated vehicle (ROV) for close-range imaging surveys of shipwreck sites. Image courtesy of Field Robotics Group, University of Michigan. Labeling was supervised by maritime archaeologists who are familiar with the sites. Labeled data allows us to train and evaluate machine learning methods for detecting shipwreck sites in side-scan sonar imagery.įigure B: Example label of side-scan sonar imagery to create a machine learning dataset for shipwreck detection. Working with maritime archaeologists at the sanctuary, we created high-resolution labels of shipwreck sites, distinguishing between background, ships, debris, and moorings. Image courtesy of Machine Learning for Automated Detection of Shipwreck Sites from Large Area Robotic Surveys. Figure A shows an example of the high-resolution side-scan sonar imagery we collected with the AUV over the Grecian site.įigure A: Grecian site imaged by the IVER autonomous underwater vehicle equipped with side-scan sonar. Surveys covered a total of 8.04 square kilometers (3.1 square miles) over the course of field trials. To collect side-scan sonar data, we deployed Michigan Technological University’s IVER-3 autonomous underwater vehicle (AUV). When the mission was complete, we moved to the next planned site. Each mission consisted of traveling to the initial starting point, deploying robotic systems, and conducting surveys in a local area. The team conducted eight shore-based day operations throughout the sanctuary on R/V Storm between May 23-June 3, 2022. The team also conducted preliminary imaging surveys for close-range inspection of shipwreck sites with robotic systems.įield expeditions were focused in Thunder Bay National Marine Sanctuary (TBNMS), which is unique in its abundance of known shipwreck sites of varying type, size, and wreck conditions. The main goal of the first year of field expeditions was to collect datasets to develop machine learning methods for shipwreck detection from sonar imagery. Shipwreck sites and exploratory sites from year 1 and year 2 are displayed. In 20, a multi-institute team of researchers conducted expeditions in Thunder Bay to do just that.įigure 1: Map of survey regions A, B, C, and D, where C and D are exploratory areas. By advancing and training the capabilities of marine robotic systems to search for and survey shipwreck sites autonomously, scientists aim to increase the efficiency and decrease the costs associated with such exploration efforts. Shipwrecks can help us better understand our past, but discovering and exploring them is expensive, time-consuming, and labor intensive. Machine Learning for Automated Detection of Shipwreck Sites from Large Area Robotic Surveysĭue to its maritime history and strategic location, Thunder Bay National Marine Sanctuary contains almost 100 known shipwreck sites, with over 100 shipwrecks still left to be found.Diversity, Equity, Inclusion, and Accessibility.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |