Abstract
Forecasting vessel flows is important to the development of intelligent transportation systems in the maritime field, as real-time and accurate traffic information has favorable potential in helping a maritime authority to alleviate congestion, mitigate emission of GHG (greenhouse gases) and enhance public safety, as well as assisting individual vessel users to plan better routes and reduce additional costs due to delays. In this paper, we propose three deep learning-based solutions to forecast the inflow and outflow of vessels within a given region, including a convolutional neural network (CNN), a long short-term memory (LSTM) network, and the integration of a bidirectional LSTM network with a CNN (BDLSTM-CNN). To apply those solutions, we first divide the given maritime region into grids, then we forecast the inflow and outflow for all the grids. Experimental results based on the real AIS (Automatic Identification System) data of marine vessels in Singapore demonstrate that the three deep learning-based solutions significantly outperform the conventional method in terms of mean absolute error and root mean square error, with the performance of the BDLSTM-CNN-based hybrid solution being the best.
Highlights
Forecasting traffic flows of vessels has been recognized as a challenging task in the maritime intelligent transportation system, since it could be affected by various complex factors [1,2]
Experimental results based on the real AIS data for a given area in Singapore show that the three deep learning based solutions significantly outperform the conventional method, in which the hybrid solution bidirectional LSTM (BDLSTM)-convolutional neural network (CNN) achieves the best performance
Since the BDLSTM-CNN based solution has the best overall performance, we look into this hybrid solution and analyze its prediction capability in a breakdown perspective
Summary
Forecasting traffic flows of vessels has been recognized as a challenging task in the maritime intelligent transportation system, since it could be affected by various complex factors [1,2]. In the second step, the ARIMA (auto-regressive integrated moving average) and WNN (wavelet neural network) were explored to predict the flows based on the low-rank and sparse components from the fist step He et al [17] proposed an improved Kalman model for short-term vessel traffic flow prediction. With respects to the BDLSTM-CNN based solution, each input will be first fed into the convolutional layers, they will go through the forward layer and backward layer of a bidirectional LSTM network As such, this hybrid solution is supposed to coherently learn the spatial and temporal dependencies pertaining to the vessel flows. Experimental results based on the real AIS data for a given area in Singapore show that the three deep learning based solutions significantly outperform the conventional method, in which the hybrid solution BDLSTM-CNN achieves the best performance.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
