Terminal homing for autonomous underwater vehicle docking

The use of docking stations for autonomous underwater vehicles (AUV) provides the ability to keep a vehicle on station, conducting missions for extended periods of time, with limited human interaction. However, the use of a docking station brings about challenges associated with terminal homing, position estimation, and vehicle control. A traditional single propeller-driven AUV must dock at a high relative approach velocity to maintain controllability, which can lead to serious damage to the AUV and the docking station. Alternatively, equipping a AUV with forward and aft pairs of horizontal and vertical cross-tunnel thrusters enables a hovering capability and allows for a slower, more deliberate approach that can help reduce potential damage during the terminal homing phase. Additionally, the commonly used ultra-short baseline (USBL) acoustic transponder attached to the docking station, which provides bearing and range measurements, can be asynchronous and sparse. The integration of these measurements into an optimal position estimation filter can potentially produce inaccuracies that are detrimental during docking operations. This thesis discusses the development of a hydrodynamic model and a filtering algorithm for position estimation for a cross tunnel thruster-enabled REMUS 100 AUV. The hydrodynamic model provides the capability of simulating vehicle docking with variable environmental effects. The filtering algorithm looks to provide an integrated solution of inertial navigation measurements and UBSL measurements to provide a more accurate vehicle location during docking operations.http://archive.org/details/terminalhomingfo1094549385Ensign, United States NavyApproved for public release; distribution is unlimited