KALMAN FILTER ON STATE VECTOR OF GPS RECEIVER

Abstract

le deflection of structure provides evidence and instruction for planning and prioritizing ,tructural inspection, rehabilitation, maintenance and repair. It helps us in monitoring repairs and reconstruction with the view of evaluating the effectiveness of maintenance, retrofit and repair works. A novel and viable method of measurement of deflection or deformation of structure, based on GPS observation has been o be is simple, economic, reliable. But GPS observations are usually frau ht with different types of errors. Of the errors, thrar'-'-Tc----'1om-.--;---'.roir associated with the observation of any particular structure is unique as it also depends on the characteristics of the system for which observation has been considered. In this thesis work, an algorithm based on Kalman filter has been developed to model error encrypted? in GPS observations for a civil structure. The algorithm has been implemented to test the vertical elevation of a cable stayed bridge under dynamic load condition. In this study, elevation of observation station has also been collected through total station and considered to be accurate for further analysis. The elevation from GPS observation has been filtered through the developed model to remove the errors. It has been found that filter values of elevation matches with those from total station. Thus, it may be concluded that the methodology may be ,extended to model random error associated with observation for any other structure by adopting suitable system specific parameters.