Decentralized model reference adaptive autopilot for underwater robot

An application of a model reference adaptive control algorithm, with the parameter adaptation has been considered. A nonlinear model with unknown nonlinearities describes the robot dynamics. Command signals are generated by an adaptive autopilot. All the signals are guaranteed to be bounded and the object state is proven to converge to a small neighborhood of the reference model state. To reduce a computational complexity, the autopilot is broken down into three subsystems, each with their own controller. The control performance of the closed-loop system is guaranteed by suitably choosing the design parameters. Quality of control is tested without and in a presence of environmental disturbances. Some simulation results are inserted to show the effectiveness of the approach.