Research In Integrated INS/GPS Filtering Method

The need of Inertial Navigation System/Global Positioning System (INS/GPS) augmentation system stems from the shortcoming of each individual technology. The stand-alone INS has superior short-term error performance, but position errors of INS are unbounded. On the other hand, long-term accuracy of GPS is good, but short-term accuracy is poor. The motivation for INS/GPS integration is to develop a navigation system that overcomes the shortcomings of each system.The primary objective of this research is to improve the in-flight alignment and positioning performance of an integrated INS/GPS during a rendezvous scenario where noise processes are not exactly known. In order to accomplish this, two mathematical models of INS/GPS integrated navigation system are designed after introducing several adaptive Kalman filters. Utilize the method, the observability analysis method of piece-wise constant system, to analyze the observability of above models and obtain the reasonable observable states.A composed correction method is proposed by the combing output correction with the backward correction. The simulation results show the composed correction is more effective than pure output correction, especially in the in-flight alignment.When the plant dynamics or noise processes are not exactly known, or the noise processes are not zero mean white noise, divergence problems will occur. In this paper, a fuzzy logic adaptive system (FLAS) is used to adjust the exponential weighting of a weighted Kalman filtering and prevent it from divergence. The simulation results show the fuzzy adaptive algorithm is robust and has high accuracy.