Model-based Predictive Control of an Unmanned Surface Vessel

L.C. McNinch, K.R. Muske, and H. Ashrafiuon (USA)


USV, Model Predictive Control


This paper discusses the application of model-based pre dictive control techniques to trajectory tracking and posi tion control of autonomous unmanned surface vessels. A discrete-time, nonlinear model predictive controller is used to determine the propulsion and steering control inputs that achieve a specified performance objective, such as mini mum tracking error or minimum time, through the solution of an on-line optimization problem that is solved at each controller sample period. The determination of these con trol inputs is subject to constraints that arise from the phys ical limits on the system, such as input saturation, and any specified limits on the state of the system. Nominal perfor mance is demonstrated using an experimental autonomous unmanned surface vessel as an example model system.

Important Links:

Go Back