Dead Reckoning of Dynamically Positioned Ships: Using an Efficient Recurrent Neural Network

Abstract

When a ship experiences a loss of position reference systems, its navigation system typically enters a mode known as dead reckoning (DR) to maintain an estimate of its position. Commercial systems perform this task using a state estimator that includes mathematical model knowledge. Such a model is nontrivial to derive and needs tuning if the vessel's dynamic properties change. To this end, we propose using machine learning to estimate the horizontal velocity of the vessel without the help of position, velocity, or acceleration sensors. A simulation study was conducted to demonstrate the ability to maintain position estimates during a Global Navigation Satellite System (GNSS) outage. Comparable performance is seen relative to the established Kalmanfilter (KF) model-based approach.