Enhancing Least Squares GNSS Positioning with 3D Mapping without Accurate Prior Knowledge
| Autor: | Paul D. Groves, Mounir Adjrad |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
050210 logistics & transportation
010504 meteorology & atmospheric sciences 3D city models Computer science 05 social sciences Aerospace Engineering Terrain Ranging 01 natural sciences Least squares Root mean square Non-line-of-sight propagation Position (vector) GNSS applications 0502 economics and business Electrical and Electronic Engineering 0105 earth and related environmental sciences Remote sensing |
| Zdroj: | Navigation. 64:75-91 |
| ISSN: | 0028-1522 |
| Popis: | GNSS positioning performance in dense urban areas is severely degraded due to the obstruction and reflection of the signals by the surrounding buildings. A basic GNSS position solution can exhibit errors of tens of meters, sometimes more. Here, 3D mapping is used to aid conventional ranging-based GNSS positioning. Terrain height-aiding contributes an additional virtual ranging measurement to the position solution. In addition, 3D city models predict Non-Line-Of-Sight (NLOS) reception over a large search area. The resulting NLOS probabilities aid a consistency checking algorithm that selects and weights the signal used for the final position solution. Iteration then refines the position solution. Practical test results demonstrate improvement in the horizontal and vertical accuracy of conventional ranging-based GNSS positioning in urban areas by a factor of 2.5 and 5, respectively. Using the new technique, the Root Mean Square (RMS) position error in dense urban areas was found to be 20.8 m horizontally and 12.2 m vertically. Copyright © 2017 Institute of Navigation |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text