Tightly Coupled Multi-GNSS Receiver Fusion for Robust Position Estimation in Urban Environments

Satellite based positioning is a key technology for many localization based driver assistance systems (ADASs) and Intelligent Transportation Systems (ITSs). Those systems are more and more migrating into the inner cities where the performance of Global Navigation Satellite Systems (GNSS) is strongly degraded. In order to mitigate the limited performance of GNSS in challenging environments new algorithmic approaches need to be investigated. The most common way is the fusion of additional vehicle sensor information. Particularly, the tightly coupled integration of odometry measurements and GNSS raw observations leads to promising results to overcome situations with degraded or corrupted GNSS measurements Pt Furthermore it was shown that setups with multiple GNSS receivers can also lead to an improved positioning accuracy in this paper the tightly coupled integration of raw observations originating from four GNSS receivers on one vehicle is evaluated. This is done under consideration of strongly multipath affected environments. In this paper several aspects concerning the design of an implementation as well as the limitations of [GRAPHICS] the usage of a straight forward Bayes filter for the tightly coupled multi antenna model are discussed.