default Towards accurate glider underwater navigation and synchronization: first approach using ADCP during CTD casts

By

Towards accurate glider underwater navigation and synchronization: first approach using ADCP during CTD casts. Sliwka, Jan; Garau, Bartolome; Borrione, Ines, Luca. CMRE-MR-2018-007. November 2018.

In order to perform most missions using unmanned vehicles, two fundamental quantities need to be known: position and time. On the surface the problem of time synchronization and positioning are usually solved using the Global Positioning System (GPS). The GPS does not however work underwater. In the absence of an absolute reference, the inaccuracies in the knowledge of time and position would increase with time until the unmanned vehicle is lost (both in space and in time). In this report, the authors deal with ways to improve the navigation and synchronization of an underwater glider. The sensor used for enhancing navigation is a small footprint Acoustic Doppler Current Profiler (ADCP) suitable to be glider-mounted. The intermediate step of the navigation algorithms is in fact to compute the ground velocity of the glider, which is the sum of its water velocity and the local sea current. The desired result is the position, meaning the integral of the ground velocity over time. This report focuses on developing, implementing and testing algorithms to estimate these two quantities thus providing a solution to the problem of navigation. The sea current estimation algorithms are based on Lowered ADCP techniques. Three techniques were implemented, one classic, one improved and one original. A model of the glider was implemented as a means to obtain the water velocity. The algorithms were tested using existing datasets. The glider model identification algorithms were tested using a dataset collected during the NREP17 sea trial. The sea current estimation algorithms were tested using a dataset collected during the NREP18 sea trial where an ADCP was lowered from the ship during several CTD casts. The synchronization is solved using an improved time-keeping hardware. The integration of this hardware is explained in more details in this report.