Design of Dynamic Ship Positioning Using Extended H∞ Filtering

Abstract

This paper is concerned with the extension of linear H∞ filter to non-linear systems with application to marine control systems. The structure of the Extended H∞ Filter (EHF) is similar to the Extended Kalman Filter (EKF) but it can account for uncertainties by exploiting the H∞ robustness properties. Since combined state and parameter estimation problems can be set up as a non-linear filtering problem, the EHF can provide both parameter and state estimation.EHF gain must be computed on-line using a local linearisation. The filter equations include the non-linear system description but the extended H∞ filter has the same type of approximations found in the EKF and hence there are similar difficulties in trying to produce stability and convergence results. The EKF has been successful in a wide range of applications. Using a marine control problem, it is shown that the EHF also has similar applications potential.