Abstract:The information-rich vessel movement data provided by the Automatic Identification System (AIS) has gained much popularity over the past decade, during which the employment of satellite-based receivers has enabled wide coverage and improved data quality. The application of AIS data has developed from simply navigation-oriented research to now include trade flow estimation, emission accounting, and vessel performance monitoring. The AIS now provides high frequency, real-time positioning and sailing patterns for almost the whole world's commercial fleet, and therefore, in combination with supplementary databases and analyses, AIS data has arguably kickstarted the era of digitisation in the shipping industry. In this study, we conduct a comprehensive review of the literature regarding AIS applications by dividing it into three development stages, namely, basic application, extended application, and advanced application. Each stage contains two to three application fields, and in total we identified seven application fields, including (1) AIS data mining, (2) navigation safety, (3) ship behaviour analysis, (4) environmental evaluation, (5) trade analysis, (6) ship and port performance, and (7) Arctic shipping. We found that the original application of AIS data to navigation safety has, with the improvement of data accessibility, evolved into diverse applications in various directions. Moreover, we summarised the major methodologies in the literature into four categories, these being (1) data processing and mining, (2) index measurement, (3) causality analysis, and (4) operational research. Undoubtedly, the applications of AIS data will be further expanded in the foreseeable future. This will not only provide a more comprehensive understanding of voyage performance and allow researchers to examine shipping market dynamics from the micro level, but also the abundance of AIS data may also open up the rather opaque aspect of how shipping companies release information to external authorities, including the International Maritime Organization, port states, scientists and researchers. It is expected that more multi-disciplinary AIS studies will emerge in the coming years. We believe that this study will shed further light on the future development of AIS studies.

Exploration of COLREG-relevant parameters from historical AIS-data

Grounding hazard considerations in evaluation of COLREGS collision avoidance algorithms

A quantitative method for the analysis of ship collision risk using AIS data

A Systematic Approach for Collision Risk Analysis Based on Ais Data

New frontiers in the risk assessment of ship collision

Autonomous Collision Avoidance at Sea: A Survey

Ais Data Based Identification Of Systematic Collision Risk For Maritime Intelligent Transport System

COLREG Scenario classification and Compliance Evaluation with temporal and multi-vessel awareness for collision avoidance systems

Stochastic COLREGs Evaluation for Safe Navigation under Uncertainty

Codification of Good Seamanship in Complex and Congested Waterways

Autonomous Navigation in Confined Waters -- A COLREGs Rule 9 Compliant Framework

An interpretable knowledge-based decision support method for ship collision avoidance using AIS data

Multi-Ship Collision Avoidance Decision-Making Based on Collision Risk Index

Developments and research directions for collision avoidance in mixed navigation environment for MASS: A systematic literature review with bibliometric analysis.

COLREG and MASS: Analytical review to identify research trends and gaps in the Development of Autonomous Collision Avoidance

A Machine Learning Method for the Evaluation of Ship Grounding Risk in Real Operational Conditions

Risk-based implementation of COLREGs for autonomous surface vehicles using deep reinforcement learning

How big data enriches maritime research – a critical review of Automatic Identification System (AIS) data applications

Towards A COLREGs Compliant Autonomous Surface Vessel in a Constrained Channel

Exploiting AIS Data for Intelligent Maritime Navigation: A Comprehensive Survey From Data to Methodology

A COLREGs-Compliant Deep Reinforcement Learning Approach