There is currently a significant amount of research being carried out in the area of GNSS positioning in difficult environments. It is well known that it is not always possible to obtain high accuracy positioning using GPS measurements in areas such as near to buildings, under dense foliage, in urban canyons, or indoors. This is due primarily to the limited availability of GPS signals, and when GPS signals are available, the level of multipath corrupting the measurement. For example, High Sensitivity GPS (HSGPS) receivers are currently able to acquire and track GPS signals close to buildings and even indoors, however typical accuracies that are achieved in such environments are of the order of 15 to 50m. Therefore, to help understand and mitigate the problems associated with high accuracy positioning in difficult environments, it is necessary to develop a way to test these sensors. The first step towards this is to develop a way in which the true coordinates of the positioning sensor can be known. If this can be achieved, more detailed information can be obtained about individual measurements such as the level of multipath that corrupts each range measurement. This information can then be used to develop new algorithms to mitigate such effects, or to develop models that can be used to characterise such errors.

Seamless centimetre-level positioning accuracy everywhere.

Seamless Positioning in All Conditions and Environments (SPACE) is a joint project undertaken by the University of Nottingham , University of Leeds , University College London, Imperial College London and a number of industrial partners, that aims to address some of the problems associated with obtaining high accuracy positioning in difficult environments. The focus of the project is the definition, and initial development of a test bed facility that will provide high accuracy positioning in difficult environments. The test bed has two primary functions: firstly, it will have the capability to act as a mobile ‘truth' source which can be used to test the performance of other positioning sensors. An example of which may be that a user will use the test bed to investigate the performance a HSGPS receiver or other indoor positioning sensor. Secondly, the test bed will provide the capability for users to implement and test new sensors and algorithms by providing an open architecture design. For example, future developments in MEMS inertial sensors could be tested by removing the standard SPACE inertial sensor suite and replacing it with the MEMS technology. Similarly, users will be able to test their own algorithms through the open interface.

There are essentially two approaches for achieving a test bed facility. The first approach is to provide a controlled environment such as a track-based system, an example of which is the SESSYL test facility at Laboratoire Central des Ponts et Chaussées . However, the track based system is restrictive since it is necessary to construct an environment around the test track to simulate varying conditions. Furthermore, it is preferable to obtain measurements in the true dynamic conditions that are likely to be experienced. This is particularly important for systems such as inertial navigation where dynamics can play a significant role in the performance of the system. Therefore, the alternative is to develop a test bed that can be used in any location. This is the aim of the current system that is being developed for the SPACE project.