Soft Computing Design of a Linear Quadratic Gaussian Controller for an Unmanned Surface Vehicle

To develop an unmanned surface vehicle (USV), several issues need to be addressed, however, the most important being the navigation, guidance and control which dictates the performance of the whole system. Herein, the control problem is addressed by designing a fuzzy logic based linear quadratic Gaussian autopilot for the Springer USV being developed at the University of Plymouth, UK. A model is first developed by the application of system identification techniques to the acquired vehicle data. Simulation results are presented to demonstrate the efficacy of the proposed approach