Implementation of the Unscented Kalman Filter and a Simple Augmentation System for GNSS SDR Receivers

R. Capua and A. Bottaro

Abstract:	The paper presents the results of the development of advanced PVT solutions for the Sogei’s GNSS SDR. Such Real-Time platform is based on an internally designed Front-End and is totally running on a General Purpose Notebook. The EKF deals with non-linear systems through process and observations functions linearization. Non-linearity errors can lead to poor Kalman filter performances or to divergence. Linearization does not allow correct state and measurements Covariance matrices propagation in presence of significant dynamic conditions. Jacobians introduce severe computational burdens and cannot be applied in the presence of discontinuities and singularities. In order to overcome such limitations, an Unscented Kalman Filter (UKF) algorithm has been developed and tested within the SDR. The UKF avoids linearization. Non-linear transformation is performed at each step on a set of deterministic sigma points extracted from the Covariance matrix, allowing prediction errors minimization. Dependence on initial conditions is reduced. Due to inherently quantized measurements and computational load constraints, SDR can highly benefit from the use of an UKF. In our implementation, a numeric efficient GPS L1 pseudoranges and Doppler measurements representation has been developed. A fine tuning phase has been carried out in mixed static and dynamic conditions. A test campaign has been carried out on a car equipped with the GNSS SDR. True trajectory was determined through RTK. An accuracy of 5m and improved robustness in case of abruptly changing routes is achieved. For SDR, a simple and inexpensive augmentation technique for medium level accuracy is advisable. A Local Augmentation based on Coordinates Corrections method has been implemented. Corrections are calculated at SDR rover side, using Reference Station measurements acquired through NTRIP and RTCM messages. An accuracy of 2 m is reached. Such a solution can pave the way for implementing reliable and low cost Augmentation systems totally based on SDR receivers.
Published in:	Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012) September 17 - 21, 2012 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	2398 - 2407
Cite this article:	Capua, R., Bottaro, A., "Implementation of the Unscented Kalman Filter and a Simple Augmentation System for GNSS SDR Receivers," Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012), Nashville, TN, September 2012, pp. 2398-2407.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In