Automatic Identification System

Automatic Identification Systems (AIS) provide crucial data for maritime surveillance, tracking vessel movements via radio or satellite transmissions. Current research focuses on improving the reliability and utility of AIS data by employing deep learning techniques, such as recurrent neural networks (RNNs), including GRUs and LSTMs, and transformer models, to address challenges like outlier detection, intentional AIS shutdowns, and data fusion with other sources (e.g., video). These advancements enhance maritime safety and security by enabling more accurate vessel tracking, improved detection of illegal activities, and more robust vessel traffic management in various environments, including open seas and inland waterways.

Papers

June 14, 2024

Outlier detection in maritime environments using AIS data and deep recurrent architectures
Constantine Maganaris, Eftychios Protopapadakis, Nikolaos Doulamis
Outlier Detection Underwater Environment Deep Recurrent Neural Network Maritime Surveillance Automatic Identification System

October 24, 2023

Detecting Intentional AIS Shutdown in Open Sea Maritime Surveillance Using Self-Supervised Deep Learning
Pierre Bernabé, Arnaud Gotlieb, Bruno Legeard, Dusica Marijan, Frank Olaf Sem-Jacobsen, Helge Spieker
Self Supervised Maritime Surveillance Automatic Identification System

February 22, 2023

Asynchronous Trajectory Matching-Based Multimodal Maritime Data Fusion for Vessel Traffic Surveillance in Inland Waterways
Yu Guo, Ryan Wen Liu, Jingxiang Qu, Yuxu Lu, Fenghua Zhu, Yisheng Lv
Vessel Detection Multimodal Data Fusion Maritime Surveillance Vessel Trajectory Inland Waterway COntinuous Trajectory Automatic Identification System

February 24, 2022

Unfolding AIS transmission behavior for vessel movement modeling on noisy data leveraging machine learning
Gabriel Spadon, Martha D. Ferreira, Amilcar Soares, Stan Matwin
Neural Network Machine Learning Deep Learning Model Recurrent Neural Network Noisy Data Vessel Trajectory Automatic Identification System

Automatic Identification System

Papers

Outlier detection in maritime environments using AIS data and deep recurrent architectures

Detecting Intentional AIS Shutdown in Open Sea Maritime Surveillance Using Self-Supervised Deep Learning

Asynchronous Trajectory Matching-Based Multimodal Maritime Data Fusion for Vessel Traffic Surveillance in Inland Waterways

Unfolding AIS transmission behavior for vessel movement modeling on noisy data leveraging machine learning