TrAISformer -- A Transformer Network with Sparse Augmented Data Representation and Cross Entropy Loss for AIS-based Vessel Trajectory Prediction
Preprint
Abstract
Vessel trajectory prediction plays a pivotal role in numerous maritime applications and services. While the Automatic Identification System (AIS) offers a rich source of information to address this task, forecasting vessel trajectory using AIS data remains challenging, even for modern machine learning techniques, because of the inherent heterogeneous and multimodal nature of motion data. In this paper, we propose a novel approach to tackle these challenges. We introduce a discrete, high-dimensional representation of AIS data and a new loss function designed to explicitly address heterogeneity and multimodality. The proposed model-referred to as TrAISformer-is a modified transformer network that extracts long-term temporal patterns in AIS vessel trajectories in the proposed enriched space to forecast the positions of vessels several hours ahead. We report experimental results on real, publicly available AIS data. TrAISformer significantly outperforms state-of-the-art methods, with an average prediction performance below 10 nautical miles up to ~10 hours.Keywords
All Related Versions
This publication has 33 references indexed in Scilit:
- Study of narrow waterways congestion based on automatic identification system (AIS) data: A case study of Houston Ship ChannelJournal of Ocean Engineering and Science, 2022
- A path planning control for a vessel dynamic positioning system based on robust adaptive fuzzy strategyAutomatika, 2022
- Ship behavior prediction via trajectory extraction-based clustering for maritime situation awarenessJournal of Ocean Engineering and Science, 2022
- An AIS-based deep learning framework for regional ship behavior predictionReliability Engineering & System Safety, 2021
- Trajectron++: Dynamically-Feasible Trajectory Forecasting with Heterogeneous Data2020
- AIS data-driven approach to estimate navigable capacity of busy waterways focusing on ships entering and leaving portOcean Engineering, 2020
- A model for vessel trajectory prediction based on long short-term memory neural networkJournal of Marine Engineering & Technology, 2019
- Ship trajectory uncertainty prediction based on a Gaussian Process modelOcean Engineering, 2019
- Study on collision avoidance in busy waterways by using AIS dataOcean Engineering, 2010
- AIS Database Powered by GIS Technology for Maritime Safety and SecurityJournal of Navigation, 2008