A Multi-task Deep Learning Architecture for Maritime Surveillance using AIS Data Streams - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

A Multi-task Deep Learning Architecture for Maritime Surveillance using AIS Data Streams

(1, 2) , (3) , (3) , (4, 1) , (1, 2)
1
2
3
4

Abstract

In a world of global trading, maritime safety, security and efficiency are crucial issues. We propose a multi-task deep learning framework for vessel monitoring using Automatic Identification System (AIS) data streams. We combine recurrent neural networks with latent variable modeling and an embedding of AIS messages to a new representation space to jointly address key issues to be dealt with when considering AIS data streams: massive amount of streaming data, noisy data and irregular time-sampling. We demonstrate the relevance of the proposed deep learning framework on real AIS datasets for a three-task setting, namely trajectory reconstruction, anomaly detection and vessel type identification.
Fichier principal
Vignette du fichier
A Multi-task Deep Learning Architecture for Maritime Surveillance using AIS Data Streams.pdf (17.02 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01808176 , version 1 (05-06-2018)
hal-01808176 , version 2 (13-06-2018)
hal-01808176 , version 3 (07-08-2018)
hal-01808176 , version 4 (15-10-2018)

Identifiers

Cite

Duong Nguyen, Rodolphe Vadaine, Guillaume Hajduch, René Garello, Ronan Fablet. A Multi-task Deep Learning Architecture for Maritime Surveillance using AIS Data Streams. 2018 IEEE International Conference on Data Science and Advanced Analytics (DSAA), Oct 2018, Turin, Italy. ⟨10.1109/DSAA.2018.00044⟩. ⟨hal-01808176v4⟩
1638 View
739 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More