Smart Technologies for Unmanned Surface Vessels : On the Path Towards Full Automation
Abstract: As for the automotive industry, large efforts are being made by industry andacademia to create autonomous ships. Te solutions for this is very technologyintense, as many building blocks, often relying on AI technology, need to worktogether to create a complete system that is safe and reliable to use. Even whenthe ships are fully unmanned, humans are still foreseen to guide the ships whenunknown situations arise. Tis will be done through teleoperation systems.In this thesis, methods are presented to enhance the capability of two building blocks that are important for autonomous ships; a positioning system, and asystem for remote supervision.Te positioning system has been constructed to not rely on GPS (Global Positioning System), as this system can be jammed or be spoofed. Instead, it usesBayesian calculations to compare the bottom depth and magnetic field measurements with known sea charts and magnetic field maps, in order to estimate theposition. State-of-the-art techniques for this method normally use low-accuracynavigation sensors and high-resolution maps. Te problem is that there are hardlyany high-resolution maps available in the world, hence we present a method of theopposite; namely using high-accuracy navigation sensors and low-resolution maps(normal sea charts). Te results from a 20h test-run gave a mean position error of10.2m, which would in most cases be accurate enough for navigation purpose.In the second building block, we investigated, how 3D and VR approachescould support the remote operation of unmanned ships with a low bandwidthconnection, by comparing respective GUIs with a Baseline GUI following the currently applied interfaces in such contexts. Our findings show, that both the 3Dand VR approaches outperform the traditional approach significantly. We foundthe 3D GUI and VR GUI users to be better at reacting to potentially dangeroussituations compared to the Baseline GUI users, and they could keep track of thesurroundings more accurately.
CLICK HERE TO DOWNLOAD THE WHOLE DISSERTATION. (in PDF format)