An autonomous underwater vehicle control with a non-regressor based algorithm

The control of autonomous underwater robotic vehicles includes many challenges. The highly nonlinear, time-varying dynamic behavior of the robot continually changes the parameters of the system model. The uncertainties in hydrodynamic coefficient make system identification difficult. Disturbances acting on the robot, such as ocean currents, continually affect the system. The paper presents a non-regressor based adaptive control scheme for use on an underwater robotic vehicle. The control scheme is non-model based and, thus, does not require any knowledge of the system. The adaptive control algorithm estimates the control gains defined by the combinations of the bounded constants of system parameter matrices. The performance of the algorithm was evaluated through simulation. Additionally, experimental data using the ODIN vehicle was analyzed.