Adaptive dynamic surface control for coordinated target tracking of autonomous surface vehicles using neural networks

This paper considers the coordinated target tracking problem of multiple autonomous surface vehicles subject to uncertainties and ocean disturbances. Motivated by the pure-pursuit guidance scheme developed for intercepting missiles and point-to-point navigation of ships, a formation control strategy is proposed by the follower to track a virtual target defined relative to the leader. The path that the virtual target traverse is not a priori. By employing the neural network and dynamic surface control techniques, a robust adaptive target tracking law is developed. The proposed controller guarantees that the target tracking errors converge to a small neighborhood of origin. The advantage of the proposed controller is that the coriolis/centripetal force, nonlinear damping, unmodeled hydrodynamics and disturbances from the environment can be compensated by on-line learning. An example is given to illustrate the efficacy of the proposed method.