Congratulations to Kevin D. Wolf of Software Logistics for leading his team to a third place victory in the Data Science track at the U.S. Navy’s HACK The MACHINE event last week in Seattle.


International regulations for preventing collisions at sea codify long-established international norms (COLREGS) for maritime navigation. However, many situations that are frequently encountered by mariners are not covered by those norms. Similar to automotive traffic laws, the safest maneuver in a given situation depends on many factors.


This challenge used data from ships on the high seas to develop algorithms to assist the Navy with preventing collisions for human-operated and autonomous vessels.  This presented the team with a complicated problem and a limited amount of time to answer this challenge.  The team’s solution, described by judges as “Simple yet Elegant”, used AIS data publicly available at with Jupyter Notebooks running on a high performance cloud VM to identify COLREG interactions from billions of points of data.


The team approached the challenge by filtering data into manageable datasets, then creating a directory structure with tracks including latitude, longitude, time, speed, and course over ground.  The data was organized into 15-minute time slices, and those time slices were further broken down into a set of cubes that made a 3D matrix with longitude, latitude and time as the axis.  A Time Series Analysis allowed the team to produce MMSI (unique ship identifiers) pairs and time stamps of potential COLREG interactions.  With those two MMSI values, the original ship tracks were obtained and the COLREG interactions analyzed. As shown in the image below, the team used ArcGIS to determine this was an Overtake COLREG interaction. Possible optimization strategies the team offered included building data staging algorithms, creating cloud-based sub-datasets, and devising a parallel process with job queuing architecture.



Pin It on Pinterest