Implementation of Near Real-Time Onboard Processing Software for a Slocum Glider Acoustic Doppler Current Profiler

Advancements in sensor and battery technology over the past two decades have facilitated the integration of complex instruments into autonomous underwater vehicles. The integration of such instruments onto gliders has enabled unprecedented data collection and expanded understanding of ocean processes. Recently, this understanding has been enhanced through the development of a software architecture called “backseat driver” on Slocum gliders, which can increase the efficiency of data collection by increasing the time the glider spends in regions of interest through adaptive mission control. A caveat to this advancement is that sensors that collect large volumes of data cannot efficiently transmit data to shore in real-time, which limits real-time applications (e.g. operational hurricane forecasting, marine mammal detection, water quality monitoring, etc.) and confines shoreside processing to post recovery. This can potentially lead to loss of data in the event of vehicle loss or missions where retrieval is not permissible, resulting in the need to develop a sensor agnostic real-time processing software and hardware architecture. Here, we detail the methodology that will be applied to test a real-time onboard processing algorithm through the use of the backseat driver software on Slocum glider MARACOOS04. An adjunct processor (Raspberry Pi 4) has been integrated into this glider with a data processing algorithm that will parse, process, and compress raw data from an acoustic Doppler current profiler.