Title: Intelligent Urban Positioning using Shadow Matching and GNSS Ranging Aided by 3D Mapping
Author(s): Mounir Adjrad, Paul D. Groves
Published in: Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2016)
September 12 - 16, 2016
Oregon Convention Center
Portland, Oregon
Pages: 534 - 553
Cite this article: Adjrad, Mounir, Groves, Paul D., "Intelligent Urban Positioning using Shadow Matching and GNSS Ranging Aided by 3D Mapping," Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2016), Portland, Oregon, September 2016, pp. 534-553.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: Conventional GNSS positioning in dense urban areas can exhibit errors of tens of meters due to blockage and reflection of signals by the surrounding buildings. Here, we present a full implementation of the intelligent urban positioning (IUP) 3D-mapping-aided (3DMA) GNSS concept. This combines conventional ranging-based GNSS positioning enhanced by 3D mapping with the GNSS shadow-matching technique. Shadow matching determines position by comparing the measured signal availability with that predicted over a grid of candidate positions using 3D mapping. Thus, IUP uses both pseudorange and signal-to-noise measurements to determine position. All algorithms incorporate terrain-height aiding and use measurements from a single epoch in time. Two different 3DMA ranging algorithms are presented, one based on least-squares estimation and the other based on computing the likelihoods of a grid of candidate position hypotheses. The likelihood-based ranging algorithm uses the same candidate position hypotheses as shadow matching and makes different assumptions about which signals are direct line-of-sight (LOS) and non-lineof-sight (NLOS) at each candidate position. Two different methods for integrating likelihood-based 3DMA ranging with shadow matching are also compared. In the positiondomain approach, separate ranging and shadow-matching position solutions are computed, then averaged using direction-dependent weighting. In the hypothesis-domain approach, the candidate position scores from the ranging and shadow matching algorithms are combined prior to extracting a joint position solution. Test data was recorded using a u-blox EVK M8T consumer-grade GNSS receiver and a HTC Nexus 9 tablet at 28 locations across two districts of London. The City of London is a traditional dense urban environment, while Canary Wharf is a modern environment. The Nexus 9 tablet data was recorded using the Android Nougat GNSS receiver interface and is representative of future smartphones. Best results were obtained using the likelihood-based 3DMA ranging algorithm and hypothesis-based integration with shadow matching. With the u-blox receiver, the single-epoch RMS horizontal (i.e., 2D) error across all sites was 4.0 m, compared to 28.2 m for conventional positioning, a factor of 7.1 improvement. Using the Nexus tablet, the intelligent urban positioning RMS error was 7.0 m, compared to 32.7 m for conventional GNSS positioning, a factor of 4.7 improvement. An analysis of processing and data requirements shows that intelligent urban positioning is practical to implement in real-time on a mobile device or a server.