Accurate estimation of vessel arrival time is crucial for optimizing port resources, such as allocating berths and dispatching hoisting machines. However, vessel passage in inland rivers is challenging due to limited space and increasing flow of vessels. Current travel time prediction models have limitations due to inadequate traffic information, especially in terms of traffic context. To address this issue, we propose a deep learning model that utilizes complex networks to describe vessel interactions’ traffic context. The model maps the correlation between various traffic factors and travel time by employing a convolutional neural network (CNN) to extract spatial trajectory features, which are merged with traffic context. A long short-term memory (LSTM) network captures temporal dependence in consecutive channel sections’ feature sets. We demonstrate the superiority of our proposed model over classical models in predicting vessel travel time, based on real data from the Wuhan section of the Yangtze River in China.
Contextualized Spatial–Temporal Network for Inland Vessel Travel Time Prediction Based on AIS data
04.08.2023
1150940 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Contextualized Spatial-Temporal Network for Taxi Origin-Destination Demand Prediction
ArXiv | 2019
|JSTC: Travel Time Prediction with a Joint Spatial-Temporal Correlation Mechanism
DOAJ | 2022
|