Enhancing Conventional GNSS Positioning with 3D Mapping without Accurate Prior Knowledge

Mounir Adjrad and Paul D. Groves

Peer Reviewed

Abstract: GNSS positioning performance in dense urban areas is severely degraded due to the obstruction and reflection of the signals by the surrounding buildings. A basic GNSS position solution can exhibit errors of tens of metres. This paper contributes towards the goal of real-time metres-level mobile positioning in outdoor urban environments by making use of spatial data in the form of 3D city models. We assess the extent to which the performance of heightaided conventional ranging-based GNSS positioning can be improved by finding the best way of determining which signals are non-line-of-sight (NLOS) or multipath contaminated. In contrast to previous approaches, we do not assume that the position is already known to within a few metres. Instead, we consider the case where the initial position, from a basic GNSS solution, is only accurate to within a few tens of metre. Therefore, which satellites are directly visible will vary across the search area. We thus use the 3D city model to predict the probability across the search area of each satellite signal being directly visible. Practical test results demonstrate improvement in the horizontal and vertical accuracy of conventional rangingbased GNSS positioning in urban areas by 58% and 78%, respectively.
Published in: Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015)
September 14 - 18, 2015
Tampa Convention Center
Tampa, Florida
Pages: 2397 - 2409
Cite this article: Adjrad, Mounir, Groves, Paul D., "Enhancing Conventional GNSS Positioning with 3D Mapping without Accurate Prior Knowledge," Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015), Tampa, Florida, September 2015, pp. 2397-2409.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In