The performance of hybrid GPS and GLONASS
In recent years, the market served by satellite positioning systems has expanded exponentially. It is stimulated by the needs of an ever increasing number and variety of scientific, business and leisure applications. The dominant system is the USA's GPS, or Global Positioning System. However, GPS is not a panacea for all positioning tasks, in any environmental situation. For example, two of the fastest growing applications, vehicle tracking and personal location, operate in an often harsh signal reception environment. This can be so severe that even with the current 29 working satellites, GPS may struggle to perform. In exceptional circumstances it can fail to provide a positioning service at all. The simplest way to improve the situation when signal reception is poor, is to add similar signals from alternative satellite systems. This has already been achieved by combining GPS with the Russian satellite positioning system, Global'naya Navigatsionnaya Sputnikova Sistema, abbreviated to GLONASS. The combination of GPS with GLONASS is referred to here as Hybrid. But how good is Hybrid relative to GPS, and how can performance be evaluated objectively? The research project presented here set out to answer this question, and to understand the situations in which Hybrid failed, and ask what solutions were then available to fulfil a positioning task. The problems associated with integrating one satellite positioning system with another, their potential inconsistencies and their impact on positioning errors were also examined. This field of research is relevant to Hybrid as defined here, and also to other mixed systems, for example GPS with EGNOS, a European geostationary satellite system, and GPS with Galileo, a proposed global system controlled by the Europeans. The issues were addressed from the viewpoint of practical usage of the positioning systems. Hence the many and varied experiments to quantify positioning performance using both static receivers, and a variety of platforms with wide ranging levels of vehicle dynamics. The capability of satellite positioning systems to work in the harshest environments, was tested in the proposed Olympic sport of bob skeleton. This involved the development of the acquisition system, and a number of programs. The latter were equally applicable to the ensuing work with road vehicles, and the quantitative assessment of positioning performance relative to a truth. The processes established to manipulate, import, and merge satellite based vehicle tracking data with Ordnance Survey digital mapping products, have already been used in four other projects within the School of Civil Engineering. The software to regularise positioning interval, smoothing processes, and to compare tracking data with a truth, have been similarly provided. Without major funding the outlook for GLONASS and hence Hybrid looks bleak, and it is predicted that without replenishment the constellation may fall to six satellites by the end of 2001. However as mentioned above, the issues identified, and ideas and software developed in this research, will be directly applicable to any future hybridisation of GPS with Galileo.