On frequently navigated waterways, ship operators and local authorities must have maps of seabed and coastline features. Newly adapted autonomous ships will soon make creating such maps easier than ever.
As it stands, well-trained technicians working on specially equipped vessels must manually map the bottom and coastal topography of harbors, lakes and rivers.
The resulting maps help prevent ships from being stranded, and they also provide insight into how various underwater and shoreline features change over time. For these reasons, it’s important to update the map regularly – the more often the better.
To help achieve this goal, scientists at Germany’s Fraunhofer Institute for Optoelectronics, Systems Technology and Image Development have significantly modified an existing commercial unmanned surface vehicle (USV), enabling it to independently perform the actual mapping of waterways .
The user first specifies the geographic area to be mapped on the software that controls the ship. The software then calculates a surface route covering the area, which the USV continues to follow automatically. It is guided along this route by GPS, accelerometers, and a sensor called a Doppler velocity log, which uses sound pulses projected downward to estimate relative speed.
Fraunhofer IOSB
The ship also utilizes a multibeam sonar system to create a 3D map of the bottom, while using waterborne lasers and cameras to create a 3D map of the coastline.Lasers and cameras are also used to spot and avoid surface obstacles, such as other ships, which will no included in the final map.
Successful tests of operational prototypes have been carried out at several different lakes.
“Currently, manual surveys are only carried out every one or two years, providing results that are far less accurate than our comprehensive 3D models, and therefore do not optimally capture the state of the watercourse,” says Dr. Janko Petereit of the Fraunhofer Institute. “Our semi-autonomous navigation system provides a cost-effective alternative to current measurement methods.”
source: Fraunhofer
