An integrated navigation, guidance and control system is designed for an uninhabited surface vehicle named Springer built at Plymouth University. Interval Kalman filtering is used for navigation. Line of sight is used for guidance. Model predictive control is utilised to build the control system, or autopilot. The performance of this innovative navigation, guidance and control system is benchmarked against a conventional optimal method based on a linear quadratic Gaussian controller which uses traditional Kalman filtering. The performance of the systems are compared and analysed in this paper.