Feature Map Aided Robust High Precision GNSS Positioning in Harsh Urban Environments

In this contribution, we propose the GNSS Feature Map-aided robust extended Kalman filter, which can provide centimeter-to-decimeter-level GNSS RTK position accuracy in urban environments without the need of additional sensors, city model information or computational intensive ray tracing methods. In this approach, the information on the predicted observation error magnitudes from the generated GNSS Feature Map is combined with the concept of robust estimation. The RTK positioning performance comparison for a dynamic experiment under harsh signal propagation conditions reveals that GNSS Feature Map-aided weighting using the Geman-McClure loss function shows the best overall performance. The RMS of the horizontal position error is improved by 17 % compared to C/N<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> weighting, while 3DMA NLOS exclusion even degrades the solution. Furthermore, the combination of Feature Map information with the robust Geman-McClure loss function is effectively enhancing the float solution and reducing the number of falsely fixed ambiguities.