Around 90% of the world trade is transported by the international shipping industry. Any disruption to world shipping lanes thus has a significant impact on the world economy. The recent steep rise of maritime piracy represents one of the biggest threats to maritime shipping in decades. Even though various countermeasures have been put into effect, no solution has been found yet, and the pirates grow stronger with each year. Only in 2010, 53 cargo-vessels were hijacked and 1181 crew members held hostage. The situation has been further deteriorating in 2011, with further steep increases in attack numbers, pirate brutality and ransom paid, which reached an all-time high in April 2011 with $13M paid for the release of the Greek owned oil tanker Irene SL.
We explore how multi-agent systems, a branch of artificial intelligence, can be used to improve maritime security, with particular focus on fighting maritime piracy. Our ultimate objective is to develop an integrated set of algorithmic techniques for maximizing transit security given the limited protection resources available. We achieve this by improving the coordination of the movement of merchant vessels and naval patrols, while taking into account the behavior of pirates.
In order to evaluate the proposed techniques and to gain better insight into the structure and dynamics of maritime piracy, we also employ agent-based simulation and machine learning techniques to build dynamic models of maritime transit and to model and assess piracy risk.
All methods are implemented within a modular software testbed featuring a scalable simulation engine, connectors to real-world data sources and powerful visualization front-end based on Google Earth.
We use various data feeds and analysis results obtained from the simulation testbed to compute the risk of attack along a route through the pirate infested areas: AgentC Online.
Michal Jakob (project leader, contact person), Ondrej Vanek, Ondrej Hrstka, Branislav Bosansky, and Michal Pechoucek (principal investigator)
The work presented is supported by the Office of Naval Research project no. N00014-09-1-0537 and by the Czech Ministry of Education, Youth and Sports under Research Programme no. MSM6840770038: Decision Making and Control for Manufacturing III.