This year at the BMDA 18 workshop Konstantinos Chatzikokolakis from MarineTraffic presented the paper “Mining Vessel Trajectory Data for Patterns of Search and Rescue”.
What is BMDA?
BMDA (Big Mobility Data Analytics) is a yearly big-data workshop that happens in the context of the broader EDBT/ICDT joint conference. Its purpose is to bring together researchers and key industry stakeholders interested in scalable data-intensive applications that manage and analyze big mobility data. The workshop focuses the exchange of ideas to solve multidisciplinary real-world problems, with a focus on innovative solutions that take advantage of emergent opportunities within the field.
This paper and the research that goes along with it has as a purpose to explore the possibility of automatically detecting Search and Rescue (SAR) activity automatically, without external information and simply relying on the ship trajectories. Using a large volume of historical AIS data, the SAR activity is detected in a scalable, data-driven, supervised way, specifically via a parallelised nonparametric statistical method (Random Forests), with accuracy rates above 77%.