Great Lake Under Ice Remotely Operated Vehicle (ROV) imagery, Near Keweenaw Waterway, Houghton, Michigan, March, 2021

Ice-water interface is an under sampled region. Instead of using invasive coring and ice-tethered instruments, we are developing navigation algorithms for unmanned underwater vehicles to survey this region autonomous with limited human intervention. For this study, we modified a Blue ROV-2 (Remotely Operated Vehicle) with additional perception sensor and navigation sensors to obtain a better localization and imaging results while surveying the ice from underneath at a standoff distance of about 2 meters. In March 2021, the ROV is deployed through an ice hole to survey the underside of the freshwater ice. The survey transect is about 50 meters long, and the ROV was remotely piloted by the operator to keep about 2 meters from the ice surface. This data set includes the results from our first field tests conducted at the Keweenaw Waterway at Houghton, Michigan in March 2021. The two video footages are recorded using two Forward-Looking Infrared (FLIR) machine vision camera at a baseline of about 9 centimeters (cm). The odometry (ROV trajectory) is processed by applying a Multi-State Constraint Kalman Filter to the measurements from the inertial measurement unit (IMU) and Doppler Velocity Log (DVL). More information about the ROV and the field test could be found in the following paper. For more information about the ROV please refer to the paper "Zhao, L., Zhou, M., Loose, B., Cousens, V.C., and Turrisi, R. (2021). Modifying an Affordable ROV for Under-ice Sensing. OCEANS 2021: San Diego – Porto, 1-5"