Wave glider dynamic modeling, parameter identification and simulation

Abstract

The wave glider is a new autonomous surface platform created by Liquid Robotics Incorporated with unique capabilities for persistent observation in a variety of marine environments. The applications of this novel technology continue to expand; some of the potential oceanographic applications require advance control and precise positioning and navigation beyond the platform's current capabilities. Having a dynamic model allows for control system development in simulation as opposed to typical trial-and-error field tuning. This also allows for high precision positioning using a number of estimation techniques that rely on a dynamic model. This thesis presents a six degree of freedom set of nonlinear dynamic equations of motion for the wave glider. This work also identifies the key hydrodynamic parameters using analytical and experimental methods. The resulting model is a reduced lumped parameter model for the purpose of control and estimation. A simulation of the model is achieved through numerical integration of the dynamic equations of motion coupled with periodic position and Euler angle models. Finally, this simulation output is then compared against a set of field trials to verify the ability of the model to emulate observed wave glider motion.