Automating Oceanography: A Robotic Surface Sensor Platform Combining Flexibility and Low-cost
This work details the design, development, and testing of the autonomous surface vessel, the SeaSlug, including both its ground control station and simulation environments. This project differs from existing commercial and research platforms as it has been specifically designed to: (i) provide sufficient payload capacity and modularity for easy modification of the system's hardware and firmware, (ii) allow for extensive out-of-water testing within a robust simulation environment, and (iii) operate in the open ocean for a variety of medium-duration missions.
The system is built around a central communications network using the CAN protocol, with components operating as modular parts of well-defined subsystems. This facilitates either the modification or addition of hardware to support a wide variety of mission types. At 6.7m in length, sufficient payload capacity is available for additional sensors in a large internal cabin and two external payload bays that run vertically through the hull. An extensive simulation environment allows for testing of all subsystems before mission deployment.
Using the L2+ guidance algorithm adapted from unmanned aerial vehicles, the SeaSlug is capable of following a desired trajectory to within 2.0m across a range of weather conditions and sea states. This has supported its use for missions that include testing solar panel integration for extended mission endurance; evaluating additional guidance, navigation, and control algorithms; and collecting oceanographic data on thermal fronts using a scientific sensor adapted to the rear payload bay.