Multi-GNSS simulations for satellite point positioning.

摘要

Global positioning and navigation relies on Global Navigation Satellite Systems (GNSSs), the best known among these being the Global Positioning System (GPS), followed by GLObal NAvigation Satellite System (GLONASS). With ongoing development of the Galileo and Compass satellite navigation systems, it appears that multi-GNSS data processing will become the norm in the next 20 years. It is now an opportune moment to evaluate and develop the potential benefits that may accrue from processing multi-GNSS data. The desire to pursue such an opportunity has given rise to the Multi-GNSS Observables Simulator (MGOS), a simulation package developed during the course of this research to study multi-GNSS scenarios involving multiple receiver platforms. A primary objective of this research is to evaluate how well point positioning can be applied to Low Earth Orbit (LEO) and Highly Elliptical Orbit (HEO) satellites. MGOS has a modular design, and provides a graphical user interface. Fundamentally, it simulates GNSS satellite and receiver trajectories, GNSS measurement noise, and GNSS receiver measurements. MGOS' simulation capabilities have been tested with external processing software for Precise Point Positioning (PPP) and relative positioning. MGOS is shown to be capable of simulating realistic measurements that allow for centimetre-level positioning or better. and its built-in own processing module is demonstrated to operate on par with high-quality Precise Point Positioning (PPP) processors.;Analysis of position errors, Geometric Dilution of Precision (GDOP), and filter performance were evaluated. Position Root Mean Squared Errors (RMSEs) (3D) for the LEO simulations are shown to range between 2.0 metres (GPS-only case) to 0.9 metres (GPS + GLONASS + Galileo + Compass case). For the HEO simulations it has been found that although it is generally beneficial to track more satellites, each new satellite increases measurement noise which can negatively affect positioning performance. The issue of diminishing returns is also investigated by systematically adding GNSSs. Positioning performance improvements diminish with each GNSS added for both LEO and HEO simulations.;In terms of the LEO and HEO studies, multiple GNSSs were incrementally combined to show how an increased number of satellites affects positioning performance. The LEO and HEO orbit types were chosen to exploit the dramatic contrast in their orbital dynamics. In addition, a sidelobe tracking feature was introduced to HEO scenarios to allow tracking of more satellites. Simulations show that multi-GNSS LEO positioning is similar in performance to kinematic ground based positioning. In the HEO case, it is shown that a multi-GNSS receiver is a necessity, since satellite geometry is very poor. Sidelobe tracking is desirable, since it can improve positioning performance to almost that afforded by LEO satellites (∼1 m or less).