Automatic Identification System (AIS) Data Supported Ship Trajectory Prediction and Analysis via a Deep Learning Model

first_pagesettingsOrder Article Reprints Open AccessArticle Automatic Identification System (AIS) Data Supported Ship Trajectory Prediction and Analysis via a Deep Learning Model by Xinqiang Chen 1,2,3ORCID,Chenxin Wei 3,Guiliang Zhou 1,*,Huafeng Wu 4ORCID,Zhongyu Wang 5 andSalvatore Antonio Biancardo 6ORCID 1 Jiangsu Key Laboratory of Traffic and Transportation Security, Huaiyin Institute of Technology, Huai’an 223003, China 2 Institute of Atmospheric Sciences, Fudan University, Shanghai 200438, China 3 Institute of Logistics Science and Engineering, Shanghai Maritime University, Shanghai 201306, China 4 Merchant Marine College, Shanghai Maritime University, Shanghai 201306, China 5 College of Transport & Communications, Shanghai Maritime University, Shanghai 201306, China 6 Department of Civil, Architectural and Environmental Engineering (DICEA), University of Naples Federico II, 80125 Naples, Italy * Author to whom correspondence should be addressed. J. Mar. Sci. Eng. 2022, 10(9), 1314; https://doi.org/10.3390/jmse10091314 Received: 17 August 2022 / Revised: 13 September 2022 / Accepted: 15 September 2022 / Published: 16 September 2022 (This article belongs to the Special Issue Marine Intelligent Transportation Systems: Data Mining and Control Optimization) Download Browse Figures Versions Notes Abstract Automatic Identification System (AIS) data-supported ship trajectory analysis consistently helps maritime regulations and practitioners make reasonable traffic controlling and management decisions. Significant attentions are paid to obtain an accurate ship trajectory by learning data feature patterns in a feedforward manner. A ship may change her moving status to avoid potential traffic accident in inland waterways, and thus, the ship trajectory variation pattern may differ from previous data samples. The study proposes a novel ship trajectory exploitation and prediction framework with the help of the bidirectional long short-term memory (LSTM) (Bi-LSTM) model, which extracts intrinsic ship trajectory features with feedforward and backward manners. We have evaluated the proposed ship trajectory performance with single and multiple ship scenarios. The indicators of mean absolute error (MAE), mean absolute percentage error (MAPE) and mean square error (MSE) suggest that the proposed Bi-LSTM model can obtained satisfied ship trajectory prediction performance.