Nonlinear filtering algorithms for GPS using pseudorange and Doppler shift measurements

This paper presents the results obtained in our research about application of modern nonlinear filtering techniques to GPS based position estimation. The stand-alone GPS based position estimation problem using GPS raw data, pseudo-range and Doppler shifts measurements are described. A new model for position and velocity estimation are then developed for nonlinear filtering. The model is nonlinear and has variable measurement number for coping with an arbitrary number of satellites. The model is investigated applying it to two different nonlinear filters. The first one is the presently most used nonlinear filter: the extended Kalman filter. The use of unscented filter as an alternative filter for composing GPS based system and model parameter learning is also proposed. The first experimental results that comprise the comparison of estimation results obtained with the filtering modal using different filters are then presented. Future research directions are also discussed.