RAIM and Complementary Kalman Filtering for GNSS Reliability Enhancement

Abtract - The objective of this research is to improve reliability and positioning accuracy of a mobile, standalone GNSS receiver in personal positioning. We propose a novel algorithm that fuses carrier information with code phase measurements and uses the additional security feature of receiver autonomous integrity monitoring (RAIM) and fault detection and exclusion (FDE) in order to detect and exclude erroneous measurements. The weighted least squares (WLS) method completed with RAIM/FDE is used to compute the GPS position and velocity estimates from pseudorange and delta range measurements. These estimates are combined in a complementary Kalman filter (CKF), which gives the velocity smoothed position estimate. The performance of the algorithm has been verified with pedestrian navigation tests; measurements for the tested algorithms were obtained using a SiRF Star II GPS receiver. Accuracy of the obtained position solutions was compared with a DGPS position solution, recorded using Thales MobileMappers. As the measure of accuracy, we used the horizontal distance of the standalone position solution from the DGPS track, i.e. the cross track error (CTE). The maximum and mean CTE values of 51.8 m and 6.3 m, respectively, were obtained with least squares (LS) position solution. With WLS and RAIM/FDE processed position, the maximum and mean errors were 25.2 m and 3.5 m. Applying CKF on WLS and RAIM/FDE processed position and velocity the values were 8.4 m and 3.1 m. Applying CKF on WLS and RAIM/FDE processed position and WLS velocity the values were 7.7 m and 2.8 m. Thus, adding the CKF to the WLS + RAIM/FDE processing reduces both the maximum and mean errors; the CKF has the largest effect on the maximum errors. Reports of this type of CKF for fusing GPS carrier and code phase information, with the additional security feature by RAIM/FDE, do not exist in the open literature.