Kalman Filters for Forecasting Open-Ocean White Shipping Location

Abstract

Merchant vessels travel across the ocean daily to deliver goods and transport cargo or passengers. Understanding the forecasted locations of these vessels is important for many reasons, including collision avoidance. Currently, their captains rely on radar, a global positioning system (GPS) satellite fix, and the Automatic Identification System (AIS) to maintain timely awareness of their surroundings. This article describes a Johns Hopkins University Applied Physics Laboratory (APL) team’s research into using a Kalman filter to improve forecasts of vessels’ locations. When provided historical geospatial data that contain uncertainties, the Kalman filter algorithm provides a means to estimate future locations of moving objects. The APL team confirmed that when using GPS and AIS data, the Kalman filter forecasting tool can predict the future location of a vessel 90% of the time within 15 nautical miles for 12 h into the future.