Conversion timing of seafarer’s decision-making for unmanned ship navigation

The aim of this study is to construct an unmanned ship swarms monitoring model to improve autonomous decision-making efficiency and safety performance of unmanned ship navigation. A framework is proposed to determine the relationship between on-board decision-making and shore side monitoring, the process of ship data detection, tracking, analysis and loss, and the application of decision-making algorithm, to discuss the different risk responses of specific unmanned ship types under various latent hazard environments, particularly in terms of precise conversion timing in switching over to remote control and full manual monitoring, to ensure safe navigation when the capability of automatic risk response inadequate. This frame-work makes it easier to train data and the adjustment for machine learning based on Bayesian risk prediction. It can be concluded that the automation level can be increased and the workload of shore-based seafarers can be reduced easily