Abstract:
In urban canyons, the positioning accuracy obtainable from global navigation satellite systems (GNSS) is mainly impaired by signal interference due to multipath and non-l...Show MoreMetadata
Abstract:
In urban canyons, the positioning accuracy obtainable from global navigation satellite systems (GNSS) is mainly impaired by signal interference due to multipath and non-lineof-sight (NLOS) effects. GNSS is one of the sensors used in connected autonomous vehicles (CAVs) for positioning, navigation and timing (PNT). Hence, it is essential that GNSS receivers in CAVs are robust and resilient. In this paper, a method consisting of two layers of GNSS observation checks is suggested to exclude these effects in order to improve the positioning accuracy. The first layer excludes all non-consistent measurements identified by a chi-square test threshold. The second layer uses a decision tree for the exclusion of any remaining multipath/NLOS affected measurements, based on a data set obtained from a ray tracer for a 3D mapped model environment. The simulation results show an enhancement in positioning accuracy greater than 95%.
Date of Conference: 20-21 August 2020
Date Added to IEEE Xplore: 29 September 2020
ISBN Information: