Levels of automation in maritime autonomous surface ships (MASS): a fuzzy logic approach

Abstract

The development of maritime autonomous surface ships (MASS) is on the verge of opening a new era in the shipping industry that can result in a new paradigm shift with regard to efficiency, safety, security, and environmental impact. This develop- ment is likely to have significant implications for cargo transportation, navigation, and ship operations. Until now, the primary research focus has been on the technical aspects of autonomous ships, with a strong focus on operational risks. There has been less attention paid to the level of automation (LOA) in which autonomous ships are expected to operate. Despite the pervasive use of LOAs, there is no agreement as to what constitutes level of automation and how LOAs in different taxonomies (as defined below) can be compared, leaving the concept prone to interpretation. This study presents the current status of levels of automation in MASS and proposes a new approach to address the shortcomings in existing LOAs (e.g., imprecision and ambiguity). A fuzzy rule-based inference system and operational criteria for auto- mation are used to quantify and express the logical sequence in the levels of automa- tion. More specifically, LOAs in MASS are defined based on LOAs in operational tasks and functions. Our approach offers a universal language to express the LOAs in MASS, thus meaningfully operationalizing an abstract concept.