Prediction of Ship Movement Using a Kalman Filter Algorithm

Abstract

The Automatic identification system is a system used globally by ships that allows the easy identification of other surface contacts (also equipped with an automatic identification system equipment). The main objective of the present work is to validate experimentally the prediction of a ship's movement. A Kalman Filter algorithm, a recursive technique widely used in several scientific areas of study including navigation, is designed to estimate and predict the movement. In this work, a non-linear kinematic model is proposed to analyse and predict the movement of an automatic identification system contact, based on the data transmitted by it. The results evidence that the automatic identification system data can be used as input of a Kalman filter to get consistent and reliable information, an advantage to decision makers so they can act with adequate knowledge and in time.