Ensemble Kalman Filter with a Square Root Scheme (EnKF-SR) for Trajectory Estimation of AUV SEGOROGENI ITS

Abstract

Results of a study on the development of navigation system and guidance for AUV are presented in this paper. The study was carried to evaluate the behavior of AUV SEGOROGENI ITS, designed with a characteristic length of 980 mm, cross-section diameter of 180 mm, for operation in a 3.0 m water depth, at a maximum forward speed of 1.94 knots. The most common problem in the development of AUVs is the limitation in the mathematical model and the restriction on the degree of freedom in simulation. In this study a model of linear system was implemented, derived from a non-linear system that is linearized utilizing the Jacobian matrix. The linear system is then implemented as a platform to estimate the trajectory. In this respect the estimation is carried out by adopting the method of Ensemble Kalman Filter Square Root (EnKF-SR). The EnKF-SR method basically is developed from EnKF at the stage of correction algorithm. The implementation of EnKF-SR on the linear model comprises of three simulations, each of which generates 100, 200 and 300 ensembles. The best simulation exhibited the error between the real tracking and the simulation in translation mode was in the order of 0.009 m/s, whereas in the rotation mode was some 0.001 rad/s. These fact indicates the accuracy of higher than 95% has been achieved.