Enhancing GNSS Positioning with 3D Mapping

The number of global navigation satellite systems (GNSS) signals that can be received in dense urban areas has increased through the availability of multiple satellite constellations and high sensitivity receivers. However, in these constrained environments, the blockage and reflection of many of the signals by buildings and other obstacles means that poor positioning accuracy remains a problem that needs to be solved. Currently, no single positioning technology can cost effectively achieve reliable real-time metres-level positioning in dense urban areas. In response to these performance requirements, a concept known as intelligent urban positioning (IUP) has been proposed whereby multi-constellation GNSS is combined with three-dimensional mapping enabling techniques such as height aiding, non-line-of-sight (NLOS) detection and shadow matching. This paper demonstrates how conventional ranging-based GNSS positioning in dense urban areas can be enhanced using fully integrated and automated terrain height aiding exploiting data from 3D mapping. Using GPS and GLONASS data collected in London, it is shown that terrain height aiding can improve the single-epoch horizontal positioning accuracy by 43%. The paper also summarises UCL’s plans for a full implementation of the IUP concept under the Intelligent Positioning in Cities project