Evaluation of Shadow Maps for Non-Line-of-Sight Detection in Urban GNSS Vehicle Localization with VANETs - The GAIN Approach

Abstract

Vehicle localization using satellite navigation systems like GPS is a convenient and achievable solution. In order to solve this problem in dense urban areas, erroneous measurements caused by non-line-of-sight situations need to be handled carefully. The purpose of this paper is to evaluate shadow maps-an efficient representation of satellite reception conditions-for reliable real-time vehicle localization in urban areas. Instead of classical digital 3D maps which need to be installed and kept up-to-date at vehicle level, shadow maps can be exchanged on demand with vehicular ad-hoc networks. The proposed approach-developed within the European research project GAIN-is proven to increase accuracy and integrity of single-frequency GPS receivers in urban areas without additional physical sensors. In addition to the positioning performance, the necessary requirements for the communication channel are derived and compared to available technologies. The evaluated results are generated from different real-world test drives and compared to a high accurate ground truth.