Performance assessment of 3D-mapping-aided GNSS part 2: Environment and mapping

Abstract

A full performance assessment of 3D-mapping-aided (3DMA) Global Navigation Satellite System (GNSS) in dense urban areas is presented. This second part of a two-part paper focusses on the effects of the surrounding environment and 3D-mapping, based on data collected in London using a u-blox EVK M8T GNSS receiver. Conventional GNSS, shadow matching, 3DMA ranging, and integrated 3DMA GNSS all perform best when the proportion of directly visible sky is high, the building height to street width ratio is low, and the average building height is below 20 m. 3DMA GNSS methods demonstrate maximum benefit at sky visibilities of 15% to 35%. All methods exhibit poorer accuracy in environments dominated by glass and steel buildings. Temporary features, such as large buses and lorries, also degrade 3DMA accuracy. Using full 3D city models gives significantly higher accuracy than simple block models, and missing buildings lead to larger positioning errors. Further enhancements to the 3DMA GNSS algorithms are recommended.