Application of Receding Horizon Kalman Filter to Underwater Navigation Systems

Abstract

In this chapter, the RHKF and its application for a velocity-aided underwater navigation system are discussed. Firstly, the limitations of the standard Kalman filter in underwater navigation systems are described. And then it is shown how the RHKF can be replaced in order to improve the position accuracy of the navigation system when the vehicle is to be operated under uncertain environments. It is shown that the RHKF is robust against environmental uncertainties by tracking only the most recent finite measurement. When a navigation system is completely observable, the RHKF is exact for noise-free systems. This deadbeat property indicates the finite convergence time and fast tracking ability of the filter. Thus the filter is appropriate for fast estimation under temporary disturbances. Based on simulations for the velocity-aided navigation system, it is then demonstrated that the RHKF guides a better and faster transient performance of the underwater navigation system compared to the standard Kalman filter.