DEVELOPMENT OF A GENERIC, STANDARDIZED INR ARCHITECTURE FOR THE CURRENT AND FUTURE GEOSTATIONARY REMOTE SENSING SATELLITES

摘要

A conceivably generic INR (Image Navigation and Registration) architecture is introduced in this paper, which is aimed to be a standardized INR design that can be used as a common framework for diverse INR systems. The proposed INR architecture has been developed by taking the COMS (Communication, Ocean and Meteorological Satellite) INR system as the baseline platform, maintaining the excellent part of it, and replacing the least square algorithm with the Kalman filter algorithm in the best attainable form in our knowledge. In this process, it has been attempted to push the envelope of designing and implementing the Kalman filter into this class of application, as maximally and efficiently as possible, to the best of our knowledge. It is believed that the proposed INR architecture can be used as a standard framework of the INR systems for both of the current and future geostationary remote sensing satellites, with the relevant adoption of modelling for the instrument payloads, related attitude sensors and the thermo-elastic distortion.