Data-driven event-triggered heading control for autonomous surface vehicles based on an adaptive observer

Abstract

This paper presents a data-driven event-triggered heading control strategy for autonomous surface vehicles (ASVs) based on an adaptive observer. First, an adaptive observer is constructed by employing heading angle of the ASV, which effectively mitigates the adverse effect deriving from measurement noise. Second, the proposed data-driven heading control strategy only utilizes the observed heading angle and rudder control input data, which does not require any ASV model parameter information. Meanwhile, to address the non-self-balancing characteristic of the ASV heading system, a historical rudder angle compensation term is incorporated into the rudder control criterion function. Third, to enhance the utilization efficiency of network resource, an event-triggered mechanism is constructed by using the measurement error and tracking error from the adaptive observer. Then, the stability of the presented data-driven event-triggered heading strategy has been rigorously verified by using the mathematical analysis. Finally, the simulation experiment validates the effectiveness of the heading control strategy.