Abstract:
The dynamics of an underwater vehicle are greatly influenced by the dynamics
of the thrusters. Precise control, for example to perform repeatable survey or coordinated
vehicle/manipulator control, should incorporate knowledge of thruster dynamic behavior.
An energy-based lumped parameter model of the nonlinear thruster dynamic response is
developed and experimentally verified using static and dynamic thruster relationships.
Three controllers to compensate for the nonlinear dynamics are designed including
analog lead compensation, model-based computed torque and adaptive sliding control
techniques. The proposed controller designs are implemented and evaluated in a hybrid,
one degree-of-freedom vehicle simulation using an actual thruster under digital control as
the actuator. Controller evaluation and comparison is based on observed vehicle tracking
performance.
The incorporation of thruster dynamics is shown to significantly improve
vehicle tracking performance. Superior, robust tracking performance with significant
model uncertainty is demonstrated in the application of the adaptive sliding control
technique. The evaluated adaptive controller structure may permit on-line adaptation to
complex hydrodynamic phenomena associated with complete vehicle/thruster
configurations such as cross-flow and mutual interference.
