Vessel Pattern Knowledge Discovery from AIS Data: A Framework for Anomaly Detection and Route Prediction

  title={Vessel Pattern Knowledge Discovery from AIS Data: A Framework for Anomaly Detection and Route Prediction},
  author={Giuliana Pallotta and Michele Vespe and Karna Bryan},
Understanding maritime traffic patterns is key to Maritime Situational Awareness applications, in particular, to classify and predict activities. Facilitated by the recent build-up of terrestrial networks and satellite constellations of Automatic Identification System (AIS) receivers, ship movement information is becoming increasingly available, both in coastal areas and open waters. The resulting amount of information is increasingly overwhelming to human operators, requiring the aid of… 
Traffic knowledge discovery from AIS data
This paper proposes an unsupervised and incremental learning approach to extract the historical traffic patterns from AIS data to provide operators with a configurable knowledge framework supporting day by day decision making and general awareness of vessel pattern of life activity.
Automatic Maritime Traffic Synthetic Route: A Framework for Route Prediction
This paper proposes a method to calculate route prediction from a synthetic route representation data once the picture of the maritime traffic is constructed and shows a good tradeoff between the predicted and the actual observed vessel routes.
Data-driven detection and context-based classification of maritime anomalies
A data-driven approach to anomaly detection, highlighting challenges specific to the maritime domain is proposed and builds on unsupervised learning techniques which provide models for normal traffic behaviour.
A distributed framework for extracting maritime traffic patterns
The proposed approach implements distributed processing on Apache Spark in order to improve processing speed and scalability and is evaluated using real-world AIS data collected from terrestrial AIS receivers, showing that the biggest part of any future vessel trajectory falls within the extracted shipping lanes.
Extracting Courses of Vessels from AIS Data and Real-Time Warning Against Off-Course
It appears, that from machine learning and natural language processing principles, a topic model might provide a novel approach for extracting implicit patterns underlying massive and unstructured collection of incoming data considered as documents.
Uncovering Hidden Concepts from AIS Data: A Network Abstraction of Maritime Traffic for Anomaly Detection
This article extends an existing network abstraction of maritime traffic that is based on nodes that correspond to naval areas of long stays or major turns for vessels and enriches them with semantic information about the different ways that vessels employ when traversing an edge, and proposes an alternative of the popular density based clustering algorithm DB-Scan.
Discovering Maritime Traffic Route from AIS network
The proposed framework of Maritime Traffic Route Discovery is evaluated on real AIS data and the experimental results shows that the proposed MTRD is able to extract the marine traffic route effectively and provides a cornerstone of maritime traffic knowledge for traffic management, anomaly detection, and conflict analysis in the future.
Local Anomaly Detection In Maritime Traffic Using Visual Analytics
This work proposes a visual analytics tool that uses spatial segmentation to divide trips into subtrajectories and score them, and displays these scores in a tabular visualization where users can rank trips by segment to find local anomalies.


Unsupervised learning of maritime traffic patterns for anomaly detection
The proposed approach only utilises AIS data, historical or real-time, and is aimed at incrementally learning motion patterns without any specific a priori contextual description.
Maritime anomaly detection and threat assessment
Five anomalous ship behaviours are outlined: deviation from standard routes, unexpected AIS activity, unexpected port arrival, close approach, and zone entry, and a process is described for determining the probability that it is anomalous.
Maritime anomaly detection using Gaussian Process active learning
This work presents a data-driven non-parametric Bayesian model, based on Gaussian Processes, to model normal shipping behaviour, learned from Automatic Identification System (AIS) data and uses an Active Learning paradigm to select an informative subsample of the data to reduce the computational complexity of training.
Automated anomaly detection processor
Development of an Automated Anomaly Detection Processor (AADP) that exploits multi-INT, multi-sensor tracking and surveillance data to rapidly identify and characterize events and/or objects of military interest, without requiring operators to specify threat behaviors or templates is discussed.
Trajectory pattern mining
This paper develops an extension of the sequential pattern mining paradigm that analyzes the trajectories of moving objects and introduces trajectory patterns as concise descriptions of frequent behaviours in terms of both space and time.
Fast Maritime Anomaly Detection using KD Tree Gaussian Processes
This work presents a state-of-the-art non-parametric regression model, based on Gaussian Processes, and is demonstrated to model normal shipping behaviour and allows a measure of normality to be calculated for each newly-observed transmission according to its speed given its current latitude and longitude.
Visualization of vessel movements
A geographical visualization to support operators of coastal surveillance systems and decision making analysts to get insights in vessel movements as an overlay on a map based on density fields derived from convolution of the dynamic vessel positions with a kernel.
Associative Learning of Vessel Motion Patterns for Maritime Situation Awareness
This paper outlines the use of neurobiologically inspired algorithms for situation awareness in the maritime domain that take real-time tracking information and learn motion pattern models on-the-fly, enabling the models to adapt well to evolving situations while maintaining high levels of performance.
Anomaly detection in the maritime domain
  • J. Roy
  • Computer Science
    SPIE Defense + Commercial Sensing
  • 2008
The anomaly detection prototype is depicted, the knowledge acquisition and elicitation session conducted to capture the know-how of the experts, the formal knowledge representation enablers and the ontology required for aspects of the maritime domain that are relevant to anomaly detection, vessels of interest, and threat analysis, the prototype high-level design and implementation on the service-oriented architecture of the CKEF are described.
Maritime multi-sensor data association based on geographic and navigational knowledge
An overview of the satellite-extended-vessel traffic service (SEV) system for in-situ and earth observation (EO) data association for cognitive data correlation concept shows the benefits brought to the resulting maritime recognised picture (RMP) in supporting decision making and situation awareness applications.