Improving depth averaged velocity measurements from Seaglider with an advanced acoustic current profiler, the Nortek AD2CP-Glider

Autonomous underwater gliders offer a unique sampling platform for ocean measurements with a variety of sensors. A fundamental problem with gliders (and other autonomous underwater vehicles) is locating measurements within the water column with reasonable horizontal and vertical accuracy. Underwater positioning systems, generally based on acoustic travel time, can provide reasonable accuracy but slow update rates because of the acoustic path length. The slow update rate is problematic when dynamic environmental conditions exist such as near the surface. These systems are also unsuited to adaptive sampling strategies or the kilometers long transect lines gliders often fly. Dead reckoning navigation is an ancient technique relying on estimates of speed and direction to propagate an initial position fix forward in time until a new position fix is obtained. Recent work between iRobot and Nortek has resulted in the integration of a next generation acoustic velocity profiler, the AD2CP-Glider, into the iRobot^® 1KA Seaglider™. The relative velocities measured by the AD2CP-Glider are used to improve dead reckoning position estimates throughout the water column and predict the location of the glider on surfacing based on an initial position fix obtained via GPS at the start of a dive cycle. Error sources, such as surface drift, are identified and corrected for when possible. Data from tests in a large lake are used to assess the validity of the dead reckoning navigation. Depth averaged velocity estimates derived from the dead reckoned navigation are compared to estimates obtained using a hydrodynamic model to predict glider velocity. The measurement based estimates are expected to perform better in more complex flows where the interaction between the flow and glider is not captured by the hydrodynamic model.