This is the first of a series of two papers that document the development in the area of vision-based navigation for unmanned underwater vehicles. In situations where the exact global position of the vehicle is not known, vision allows the relative position to be measured precisely to control the vehicle. The first paper deals with tracking stationary objects such as pipelines and cables, and also of moving targets such as fish. The second paper, to be published in IMarEST Proceedings Part B8, covers the topics of station-keeping, mosaicking and positioning. The most significant of the previous efforts at each are examined, including an overview of the image processing techniques used, the various motion estimation algorithms employed and the different styles of approach. Both monocular and binocular vision is examined and the outstanding challenges still to be solved are presented. A description of the most recent style of approach and a projection of how vision may help facilitate the future promise of unmanned underwater vehicles is offered.
展开▼
