Event-triggered path tracking control with obstacle avoidance for underactuated surface vessel compliant with COLREGs-constraints: Theory and experiments

This paper addresses the path tracking problem for an underactuated surface vessel with model parametric uncertainties, in the presence of unknown ocean currents under input saturation and limited transmission resources, while guaranteeing no collision can occur with a nearby target ship (TS) in compliance with the International Regulations for Preventing Collisions at Sea (COLREGs) rules. In view of COLREGs practice, this paper proposes a strategy to address the issue of collision avoidance in the presence of an encounter situation. Based on different stages of encounter situation, a local path re-planning-based repulsive potential function technique is proposed, such that the potential force generated by a decision-making algorithm, is directly acted on the original desired trajectory deviating thus the Own Ship (OS) to take direction in compliance with the COLREG constraints. By restoring to the path-tracking framework, this paper proposes a novel navigation control algorithm with ship-to-ship collision avoidance. At the navigation level, a finite-time prescribed performance control strategy with the deployment of novel asymmetrical envelopes is proposed to prescribe path-tracking errors within user-defined constraints while reducing transient response overshoots. Furthermore, in order to enhance the tracking performance under model uncertainties and external disturbances, a radial basis function neural network (RBFNN) technique is used. We first propose a second-order neural network disturbance observer to restrain the compound disturbances which are combined with external disturbances and neural network approximation errors. Next, based on the estimated compound disturbances, an adaptive NN controller is designed via an event-triggering mechanism along with a transverse function approach and the backstepping technique. It is shown that with the proposed technique, all signals in closed-loop systems are semi-globally uniformly bounded, while the output tracking errors converge to a prescribed arbitrarily small region within a finite time. In addition, we prove via the direct Lyapunov approach, the existence of minimal inter-event time and thus Zeno-behavior is avoided. Finally, experiments are conducted to validate the effectiveness and performances of the proposed control strategy.