Approaches to evaluate and improve short-term relative accuracy of GPS/INS systems

Abstract

GPS/INS integrated systems can provide accurate and reliable navigation information using data fusing algorithm such as a Kalman filter. Current research works related to GPS/INS navigation systems mainly focused on the absolute accuracy with dominant components of mid-term and long-term errors. However, the short-term relative accuracy will be required particularly in some applications, such as POS for LiDAR and InSAR, GPS/INS deeply-couple. This paper raises the importance of the short-term relative accuracy of the GPS/INS systems for certain applications. Then the method to evaluate the short-term navigation errors using Allan Variance analysis is proposed and the effort to improve the short-term accuracy by adjusting the Kalman filtering parameters is made. The feasibility and effectiveness of the proposed methods were verified by simulation results and typical field test results. The outcome of this paper will help to optimize the design of GPS/INS systems for specific applications areas where short-term relative accuracy is the key requirement; and facilitate the use of the low-end GPS/INS systems in these areas.