Inertial navigation aided with GPS information

Abstract

This paper presents a dynamic alignment algorithm for a six-degree of freedom inertial unit. A differential GPS is used as external sensor. It provides decorrelated range position and Doppler velocity information. A simplified error model valid for a local area is also presented. An indirect Kalman filter approach is used to fuse high frequency inertial information with low frequency GPS data. Experimental results are presented showing that the filter is able to predict high frequency manoeuvres as well as detect multipath errors in the GPS information.