Control Seminar
Real-Time Visual SLAM for Autonomous Underwater Hull Inspection
This talk covers the development of a real-time monocular visual simultaneous localization and mapping (SLAM) algorithm and its results for its application in the area of autonomous underwater ship hull inspection. The goal of this work is to automatically map and navigate the underwater surface area of a ship hull for foreign object detection and maintenance inspection tasks. The proposed algorithm overcomes some of the specific challenges associated with underwater visual SLAM, namely limited field of view imagery and feature-poor regions. It does so by exploiting our SLAM navigation prior within the image registration pipeline and by being selective about which imagery is considered informative in terms of our visual SLAM map. A novel online bag-of-words measure for intra- and inter-image saliency is introduced, and is shown to be useful for both key-frame selection and for information-gain based link hypothesis. An online planning framework that computes the most informative regions of the hull for revisitation in the SLAM graph, based upon visually plausible expected information-gain, is also shown. Results from several real-world hull inspection experiments are shown to validate the overall approach"”including one survey comprising a 3.4 hour / 2.7 km long trajectory.
Ryan M. Eustice received the B.S. degree in mechanical engineering from Michigan State University, East Lansing, MI in 1998, and the Ph.D. degree in ocean engineering from the Massachusetts Institute of Technology/Woods Hole Oceanographic Institution Joint Program, Woods Hole, MA, in 2005. Currently, he is an Assistant Professor with the Department of Naval Architecture and Marine Engineering, University of Michigan, Ann Arbor, with courtesy appointments in the Department of Electrical Engineering and Computer Science, and in the Department of Mechanical Engineering. His research interests include autonomous navigation and mapping, computer vision and image processing, mobile robotics, and autonomous underwater vehicles.