I-Corps: Translation Potential of an Artificial Intelligence (AI)-Enabled Autonomous Subsea Connection System

Award ID: 2630730

This I-Corps project is based on the development of an autonomous system that connects and disconnects underwater equipment without human divers. Currently, human divers or expert-piloted underwater vehicles are responsible for installing and servicing offshore devices. Commercial diving is one of the world's most dangerous professions. Robotic alternatives keep people out of the water but are hard to control precisely, since wave motion constantly shifts the vehicle, leaving no reliable way to confirm a connection has been made. In addition, commissioning a device by hand can require two dive teams, while a connection that cannot be remade in place turns even a small repair into a costly full recovery. This technology combines a self-aligning connector with artificial intelligence (AI)-enabled autonomy software that completes and verifies the connection, greatly reducing the cost. This solution may make operations and maintenance safer and more affordable across offshore energy, seafloor cabling, sensor deployments, aquaculture, and other subsea infrastructure applications. This I-Corps project utilizes experiential learning coupled with first-hand investigation of the industry ecosystem to assess the translation potential of an autonomous subsea connection system with a self-aligning interface and an artificial intelligence (AI)-enabled autonomy stack. This technology uses magnetic geometry, which are permanent magnets in paired mating modules that passively guide connector halves into alignment, relaxing the mechanical tolerances that require precise vehicle piloting. The autonomy stack weighs magnet forces against hydrodynamic disturbances from drag, currents, and waves to select the safest connection strategy using advanced AI algorithms. It then plans, executes, and verifies the connection using a constrained motion planner and controller. This stack also draws on failure-recovery methods from robotic manipulation research, adapted to the underwater domain where failed attempts are costly in vessel and support time, turning a task that demands teleoperation into one a standard autonomous vehicle can complete. Previous research has demonstrated core alignment and mating in controlled testing. In addition, the system pairs with the user's own underwater vehicle, gaining faster, safer connections without needing piloting skills or custom technology specific to each connector type, depth, and application. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Ruth Shuman