ION GNSS 2010
Session A3: GNSS Algorithms and Methods 1

Title: Wide Area RTK Augmentation Approach Based on Vehicle-borne GNSS Data
Author(s): L. Zhou, D. Huang, Southwest Jiaotong University, China; K. Zhang, S. Wu, RMIT University, Australia

The sparse distribution of the continuously operating reference stations (CORS) in most parts of the globe brought a great challenge for high-accuracy real-time kinematic seamless positioning. The reason is that the spatial resolution of ionosphere model based on sparse CORS is low. The model cannot accurately constrain the ionosphere delay and the ambiguities cannot be resolved rapidly on the rover. The ionosphere tomography technique can improve the accuracy of the model. In order to improve ionosphere model, the adequate spatial resolution of ionosphere tomography is necessary. But ionosphere is huge and electron density is non-uniform, an adequate spatial resolution means large numbers of unknown parameters and a few observations. It is easy to be an ill-conditioning problem and small input errors can lead to large perturbations in the solution. The essential method is to increase the number of observations. With the development of economy, many vehicles will be equipped with GNSS receivers and vehicle-borne GNSS observations will be an important source of real-time data.

Consequently, the current CORS networks will be expanded to large-scale dynamic networks, and the spatial and temporal resolutions of GNSS observations will be significantly improved. If a great number of vehicle-borne GNSS receivers move back and forth along fixed rails and roads in the area covered by a CORS network. The determination of these rails and roads can be more reliable and accurate via multi-temporal data analysis. These results can then be used as the constraints on the trajectories of vehicle-borne receivers so that vehicle-borne GNSS receivers and CORS form a high spatial resolution ionosphere observation network. A more accurate atmosphere delay model with high spatial and temporal resolutions can be obtained. Using this atmospheric model, precise real-time seamless navigation and positioning will be made possible. With the trend of growth and periodic change in the number of vehicle-borne GNSS receivers, the amount of data and computation will increase rapidly and change dramatically. As a result, a single server can not undertake these tasks. In order to address this issue, the cloud computing technology can be considered as a solution as this technology can provide on-demand services with scalable approaches. It can manage and schedule computing resources online, and it can also effectively store massive data and allocate massive computing tasks. This project is to improve the network RTK theory by developing a new algorithm that uses data not only from the CORS networks but also from vehicle-borne GNSS receivers and to use the cloud computing technology to implement the proposed approach. This project has significance as data from various regional networks and from different observing modes or status (i.e., static and kinematic) are integrated, which will significantly improve the positioning accuracy and service coverage of current GNSS network RTK.

Return to Session A3