Agenda

Session 1

Chair: Elena Camossi

14:00 Welcome and introduction

14:05 Anita Graser, Melitta Dragaschnig, Peter Widhalm, Hannes Koller and Norbert Braendle

Exploratory Trajectory Analysis for Massive Historical AIS Datasets

Data exploration is an essential task for gaining an understanding of the potential and limitations of novel datasets. This paper discusses the challenges related to exploring large AIS datasets. We address these challenges using trajectory-based analysis approaches implemented in distributed computing environments using Spark and GeoMesa. This approach enables the exploration of datasets that are too big to handle within conventional spatial database systems. We demonstrate our approach using a case study of 4 billion AIS records.

14:20 Gianandrea Mannarini, Lorenzo Carelli and Amal Salhi

EU-MRV: an analysis of 2018's Ro-Pax CO2 data

The EU-MRV dataset containing CO2 emission data from vessels calling at European harbours is analysed. The data is provided in aggregated form (on annual basis, for year 2018) and our focus is on ferry boats. Vessel data from: the Copernicus Marine Environment Monitoring Service, the IMO Global Integrated Shipping Information System dataset, and an open repository of ferry data are used for augmenting the original dataset. From an analysis of several efficiency indicators, some level of clustering in the vessel population is observed, with year of build, vessel length, service speed, and fuel type allowing for specific insights. Georeferencing data offers further clues on continental patterns of the Ro-Pax emissions.

14:35 Panagiotis Tampakis, Eva Chondrodima, Aggelos Pikrakis, Yannis Theodoridis, Kostis Pristouris, Haris Nakos, Eleni Petra, Theodore Dalamagas, Andreas Kandiros,

Georgios Markakis, Irida Maina and Stefanos Kavadas

Sea Area Monitoring and Analysis of Fishing Vessels Activity: The i4sea Big Data Platform

The i4sea research project provides effective and efficient big data integration, processing and analysis technologies to deliver both real-time and historical operational snapshots of fishing vessels activity in national sea areas. This paper presents the architecture of the i4sea big data platform for sea area monitoring and analysis of fishing vessels activity and demonstrates the operation of some use-case pilot scenarios.

14:50 Break

Keynote presentation

Chair: Christophe Claramunt

15:00 Jean-Claude Thill, Department of Geography & Earth Sciences at University of North Carolina at Charlotte

Structures in Networks of International Maritime Cargo Flows, by Jean-Claude Thill and Paul Jung

Maritime trade and shipping is increasingly well documented by various data sources that trace the origin and destination of each unit of shipment (like bill of lading) as well as a variety of attributes (such as the shipper, carrier, voyage, commodity type, value, and so on). Techniques of flow analysis and dimension reduction have been applied to find patterns in the data. This talk will specifically discuss the contribution of network science to extract structures hidden in the data. Specifically, we will articulate the objective of such data analytic operation, and discuss various approaches available. A particular emphasis will be on the new method of nonparametric weighted stochastic block modeling (npWSBM). The presentation will be illustrated by several use cases of analysis of U.S.-bound cargo flows from several world regions.

15:45 Break

Session 2

Chair: Laurent Etienne

16:00 Christiaan Burger, Trienko Grobler and Waldo Kleynhans

Discrete Kalman Filter and Linear Regression comparison for vessel coordinate prediction

Automatic Identification System (AIS) is one of the most prominent systems for monitoring vessel activity. Although significant advances in AIS coverage have been achieved in recent times, a vessel will typically still experience gaps in reception within its voyage. These gaps can vary from a few seconds to multiple hours depending on the route. A typical approach when dealing with these gaps when monitoring a vessel’s voyage is to move from a pointal domain to a trajectory domain using a trajectory prediction algorithm. In this paper, we compare the performance of two trajectory prediction algorithms being (1) the Discrete Kalman filter (DKF) and (2) the Linear Regression Model (LRM). It is postulated that the DKF, with its added complexities in initiating the relevant parameters does not yield significant performance advantages over using a computationally simpler LRM.

16:15 Giannis Fikioris, Kostas Patroumpas and Alexander Artikis

Optimizing Vessel Trajectory Compression

In previous work we introduced a trajectory detection module that can provide summarized representations of vessel trajectories by consuming AIS positional messages online. This methodology can provide reliable trajectory synopses with little deviations from the original course by discarding at least 70% of the raw data as redundant. However, such trajectory compression is very sensitive to parametrization. In this paper, our goal is to fine-tune the selection of these parameter values. We take into account the type of each vessel in order to provide a suitable configuration that can yield improved trajectory synopses, both in terms of approximation error and compression ratio. Furthermore, we employ a genetic algorithm converging to a suitable configuration per vessel type. Our tests against a publicly available AIS dataset have shown that compression efficiency is comparable or even better than the one with default parametrization without resorting to a laborious data inspection.

16:30 Dejan Štepec, Tomaž Martinčič, Fabrice Klein, Daniel Vladušič and Joao Pita Costa

Machine Learning based System for Vessel Turnaround Time Prediction

In this paper we present a novel system for predicting vessel turnaround time, based on machine learning and standardized port call data. We also investigate the use of specific external maritime big data, to enhance the accuracy of the available data and improve the performance of the developed system. An extensive evaluation is performed in Port of Bordeaux, where we report the results on more than 10 years of historical port call data and provide verification on live, operational data from the port. The proposed automated data-driven turnaround time prediction system is able to perform with increased accuracy, in comparison with current manual expert-based system in Port of Bordeaux.

16:45 Break

Session 3

Chair: Konstantina Bereta

17:00 Valerio Fontana, José Manuel Delgado Blasco, Andrea Cavallini, Nicola Lorusso, Alessandro Scremin and Ciro Manzo

Artificial intelligence technologies for Maritime Surveillance applications

The use of AI methods are currently evolving tasks done in the past by analyst. During the last years, technology had helped to jump into fully automatic methods for monitoring and surveillance tasks, such as object detection, change detection and many more. In this work we want to show some of the AI based models which RHEA Group has been working on which can be applied to the maritime domain, such as ship detection and vessel identification, SAR and AIS data fusion and superresolution of satellite data. Each of these models can be further extended and specialized into specific monitoring and surveillance tasks, from detection of ghost ships, measure environmental damage or monitoring of critical infrastructure near harbor or protected areas. In this paper, we illustrate some examples of the status of our research activities and the developments in these prototype applications.

17:15 Alexandros Troupiotis-Kapeliaris, Konstantinos Chatzikokolakis, Dimitris Zissis and Elias Alevizos

Experimental Comparison of Complex Event Processing Systems in the Maritime Domain

Complex Event Processing (CEP) ’s main purpose is recognizing interesting phenomena upon streams of data. So its only natural that it would find applications in the maritime domain, where detecting vessel activity plays an important role in monitoring movement at sea. In this study we briefly examine the field of Complex Event Processing; we present two CEP implementations, one based on machine learning techniques and a rule-based system modelled with Event Calculus. Finally, we evaluate their ability in modeling activities that involve multiple vessels, by comparing their results on real-life examples.

17:30 Duong Nguyen, Matthieu Simonin, Guillaume Hajduch, Rodolphe Vadaine, Cedric Tedeschi and Ronan Fablet

Detection of Abnormal Vessel Behaviors from AIS data using GeoTrackNet: from the Laboratory to the Ocean

The constant growth of maritime traffic leads to the need of automatic anomaly detection, which has been attracting great research attention. Information provided by AIS (Automatic Identification System) data, together with recent outstanding progresses of deep learning, make vessel monitoring using neural networks (NNs) a very promising approach. The global of this paper is to further analyse a novel neural network we have recently introduced —GeoTrackNet— regarding operational contexts. Especially, we aim to evaluate (i) the relevance of the abnormal behaviours detected by GeoTrackNet with respect to expert interpretations, (ii) the extent to which GeoTrackNet may process AIS data streams in real time. We report experiments showing the high potential to meet operational level of the model.

17:45 Wrap-up and closure