Trajectory Estimation of Autonomous Surface Vehicle Using Extended Kalman Filter

Abstract

Autonomous Surface Vehicle (ASV) is a ship that can move autonomously in the water surface. In this work, we focus on Touristant ASV with the following specifications: length 4 meters, diameter 1.5 meters and height 1.3 meters. The main objective of this work is estimating the position of ASV based on the linear ASV model under the influence of wind speed and wave height, where the linear model is obtained from the linearization of the original nonlinear ASV model. In this work, we apply the Extended Kalman Filter (EKF) method to the linear ASV model in order to produce a small error in position. There are 2 simulations for the implementation of the EKF method, with 200 and 300 iterations. The error of position resulting from the simulation shows that the estimation accuracy of position is 95%-98% where the error of position in the x-Axis is 0.01 meters, the error of position in the y-Axis is 0.012 meters and the error of position in the plane XY is 0.011 meters.