Urban navigation using 3D mapping-aided (3DMA) GNSS

Abstract

Localization and navigation rely on a global navigation satellite system (GNSS) to provide an absolute position. However, signal reflection and attenuation result in positioning performance degradation, which is common in urban canyons. Urban localization can be enhanced using the resources associated with 3D building models. Different 3D mapping-aided (3DMA) GNSS algorithms have been proposed, in which the 3D building models are used to aid the positioning. Recently, the candidate-based 3DMA GNSS framework was presented to examine the distributed particles, and the particles that best match the observed measurements, that is, the particles with the minimum cost, are identified as the receiver location. Such particle sampling approaches are inexpensive but incur a high computational load. Besides, the candidate-based 3DMA GNSS cannot provide reliable solutions for continuous positioning. This study develops a novel 3DMA GNSS positioning framework for urban positioning. Based on the existing candidate-based 3DMA GNSS, the candidates are distributed effectively with a context-based algorithm. Furthermore, this study harnesses the benefits brought by the complementary nature of the different 3DMA GNSS algorithms and provides an integrated solution to maximize positioning performance. The third contribution of this study is to increase the robustness of the positioning by connecting multi-epoch solutions. This study achieves this by mathematically expressing the candidate-based 3DMA GNSS. The mathematically expressed model is then tightly integrated with Doppler measurements using factor graph optimization (FGO). Designed experiments are conducted to validate the performance of the proposed positioning framework. The evaluation shows that the proposed positioning framework outperforms the conventional positioning methods. The proposed framework can provide accurate and reliable positioning for smartphones and consumer-grade GNSS receivers in urban canyons.