MARITIME SECURITY AND ANOMALY DETECTION

Current Status & Challenges
Maritime Domain Awareness: Maritime Domain Awareness (MDA) is the effective understanding of activities, events and threats in the maritime environment that could impact global safety, security, economic activity or the environment. Recent advancements in Information and Communications Technologies (ICT) have created opportunities for increasing MDA, through better monitoring and understanding of vessel movements. The Automatic Identification System (AIS) was developed primarily as a tool for maritime safety and vessel collision avoidance and is an integral component of various Vessel Traffic Services (VTS), Vessel Traffic Management Systems (VTMS) and Vessel Traffic Monitoring & Information Systems (VTMIS). EU wide, various funded research projects have been deployed in an attempt to address safety at seas through ICT, including Maritime Navigation and Information Services (MarNIS), Motorways & Electronic Navigation by Intelligence at Sea (MONALISA), Advanced National Networks for Administrations (ANNA), Vessel traffic monitoring in EU waters (SafeSeaNet) and MOS. A number of vessel tracking systems are accessible to the public through the Internet. As such, MarineTraffic (MT) (by EXMILE) is part of an open, community-based project that provides real-time information to the public regarding vessel movements and port traffic across the coast-lines of many countries around the world. MT is one of the most popular vessel tracking services, providing information on more than 460,000 vessels and processing more than 50 million position reports per day, covering more than 10,000 ports and marinas across the globe. Information regarding these vessels is collected from over 1600 AIS receivers located at the coastlines of more than 150 countries. The challenge: while in the past, surveillance had suffered from a lack of data, current tracking technologies have transformed the problem into one of an overabundance of information, leading to a need for automated analysis. The major challenge faced today by the security domain is developing the ability to identify the patterns that emerge from huge amounts of data, fused from various sources and generated from monitoring thousands
of vessels a day, so as to act proactively to minimize the impact of possible threats.

Maritime security and anomaly detection

To get a clear picture of the complex maritime environment, you cannot simply add up and connect together various vessel positions as they travel across the seas. Achieving situational awareness, perceiving and comprehending elements and their contextual meaning in the environment within a given volume of time and space, while projecting their status into a future timeframe, is a critical element of Maritime Domain Awareness. Information fusion from various sources has the capability to support the identification of inferences between objects and their environment. Combining cross-sectoral data, such as weather information, nautical information, and incident reports, with expertise can assist in improving the
understanding of a current complex environment, but also aid in attempting to predict its future state. As such, data fusion can enable improved assessment of situations and, therefore, a better and more effective response. The development of global ship tracking systems opens up possibilities of advancing maritime security far beyond simple collision prevention. Anomaly detection can be defined as a method that supports the situational-assessment process by indicating objects and situations that, in some sense, deviate from the expected, known or “normal” behaviour and thus may be of interest for further investigation. Vessel behaviour can be defined as the sum of all characteristics defining vessels movement, such as vessel position, course, heading and speed, observed over a given period of time. By definition, a pattern is composed of recurring events that repeat in a predictable manner. Vessel behaviour monitored over a long period of time can give insights into the navigational patterns followed by each vessel on specific routes. The information fused in the previous phase provides system capabilities that support the assessment of the relationships between objects (e.g. vessels) and their given environment within a given amount of time. Impact assessment processes deal with the predictions about possible future situations to determine potential impacts. Machine learning techniques can be employed to “learn” vessel behaviours and project their status into a future time frame. Information fused from a variety of resources needs to be combined into a semantically rich representation of vessel trajectories.

BigDataOcean Solution
Deviations from “normal” behaviour may be a response to environmental conditions (weather conditions, busy sea lanes/ports, etc.), in which case the system should support rerouting nearby vessels, but could also be threats related to terrorism, illegal trafficking, fishing, piracy and others. This pilot will build on previously developed capabilities so as to be able to identify such “anomalies”, classify them, create alerts for careful assessment, but also propose measures of hazard avoidance. Anomaly detection can be provided both for the current time frame but also future projected behaviour. Each event prediction will be accompanied by a risk assessment regarding collisions and environmental impact. The current times of events will be recorded and used as an input in a continuous learning procedure. In the case of a hazardous prediction, measures are taken and alternative routes are proposed. No personal data will be provided or analysed during this pilot, as the real-time data to be analysed will not contain any kind of data relating to individuals.


Comments are closed