•  
  •  
 

Abstract

The prediction of ship navigation trajectories is crucial for ensuring maritime navigation safety and forecasting risk. In recent years, extensive research has been conducted on the prediction of ship navigation trajectories at home and abroad, and some results have been achieved. At present, most research on ship navigation trajectory prediction relies heavily on automatic identification system (AIS) data and the historical trajectories of ships. This study first reviews the research background and current situation of ship navigation trajectory prediction. Second, it provides an overview of ship navigation trajectory prediction, with an emphasis on sorting out and elaborating on artificial intelligence research that methods have been popular in recent years, including machine learning algorithms, deep learning algorithms, and hybrid algorithms and their variants. It integrates and summarizes the evaluation indicators used in the relevant research, and discusses and compares the advantages and disadvantages of each artificial intelligence algorithm with regard to the prediction of ship navigation trajectories. Finally, a summary analysis is performed, the future development direction of ship navigation trajectory prediction is explored, and the gap between current academic research and practical applications, as well as overlooked issues, is identified.

Download
COinS