ABSTRACT
ABSTRACT: Satellite Systems R&D at the Institute of Engineering Surveying and Space Geodesy (IESSG) dates back to the 1970s. The use of Global Navigation Satellite Systems (GNSS), particularly the Global Positioning System (GPS) for bridge deformation monitoring started in the early 1990s and still forms a major theme of research activities at the IESSG. In recent years, joint studies with Brunel University in using GNSS for monitoring the deflection of the Humber Bridge and the London Millennium Bridge, and computational tools such as Finite Element (FE) models for the predictions of structural dynamics have been carried out, aiming at a better understanding of structural performance under a variety of active loadings such as wind, traffic, temperature and tidal current. A recent project, “A Remote Bridge Health Monitoring System Using Computational Simulation and GPS Sensor Data” has been conducted in collaboration with Cranfield University and industrial partners such as Leica Geosystems Ltd, W S Atkins, Pell Frischmann and Railtrack (now Network Rail). The overall objective of the project was the creation of a system employing advanced computational tools such as FE model coupled with GPS, triaxial accelerometers, pseudosatellites (pseudolites) and other sensors to remotely monitor the health condition of operational bridges without the need for on-site inspection. This paper addresses the limitations of GPS based bridge deformation monitoring systems according to the knowledge acquired by the authors in their practice of more than 15 years. Relevant solutions are recommended to help different users fully understand the advantages and disadvantages of this modern positioning technology. The paper is organized as follows. The first section is a brief introduction and is followed by a discussion section of the major limitations of GPS for bridge deformation monitoring. The section Solutions for Tackling Problems discusses the potential improvements of the GPS based monitoring system performance through different approaches such as filtering techniques and hybrid sensor systems. The subsequent sections are some suggestions for future work and the paper conclusions.
