10. Unmanned Surface Vessels

Autonomous navigation of unmanned surface vessels (USVs) (such as ships, hovercrafts, etc), is a significant topic, since it can find use in both security and defence tasks, as well as in maritime transportation. The problem of control and trajectory tracking for unmanned surface vessels (of the ship or hovercraft type) is non-trivial because the associated dynamic and kinematic models are complex nonlinear ones. A first problem that arises in controller design for unmanned surface vessels is that trajectory tracking has to be achieved despite modelling uncertainty and external perturbations and thus the control loop must exhibit sufficient robustness. Another problem that has to be dealt with is that the vessels model is often underactuated (the propulsion system consists of less actuators than the vessel’s degrees of freedom). The present chapter treats the problem of control of unmanned surface vessels. Solution to the associated control problem is provided through (i) global linearization methods, (ii) approximate linearization methods and (iii) Lyapunov methods. To solve the control problem for unmanned surface vessels without prior knowledge of the associated dynamic model, elaborated real-time estimation methods are developed. These allow for identifying the unknown dynamic model of the vessel and for implementing an indirect adaptive control scheme. Moreover, for the accurate localization of the vessel and for precise computation of its motion characteristics advanced (and precisely validated) nonlinear filtering and distributed filtering are applied. These enable to perform fusion of the measurements of heterogeneous sensors and of state estimates provided by individual distributed local filters. In particular, the chapter treats the following issues: (a) Nonlinear control and Kalman Filtering for a 3-DOF surface vessel, (b) Flatness-based control for the autonomous hovercraft (c) Nonlinear optimal control for autonomous navigation of unmanned surface vessels, and (d) validation of distributed Kalman Filtering for ship tracking applications.