In this dissertation, we aim to extend the development of underwater robot technologies by implementing robotics vehicles and applying Simultaneous Localization and Mapping (SLAM) approaches towards autonomous navigation. Firstly, we present the implementation of a Remotely Operated Vehicle (ROV) and a robot development framework used to upgrade two underwater robots. We then review SLAM algorithms in underwater environments and focus on visual SLAM based on monocular cameras. Finally, we apply a feature detector and graph optimization SLAM algorithm and present the results and challenges for its application.
|Qualification||Doctor of Philosophy|
|Award date||25 Feb 2019|
|Publication status||Unpublished - 2019|