Tightly Coupled Integration of a Low Cost MEMS-INS/GPS System using Adaptive Kalman Filtering

Abstract

The integration of Inertial Navigation System (INS) and Global Positioning System (GPS) can produce accurate results if four or more GPS satellites are tracked. However, in GPS attenuated signal environment the errors of a low accuracy MEMS/GPS system rapidly grow to the unacceptable level. A tightly coupled integration scheme is utilized to improve the performance and reliability of the low accuracy integrated system in the areas such as tunnels, tall buildings, urban canyon, and forest canopy. This model is capable to detect the GPS fault and to track the errors of the integrated system even when less than four satellites are being tracked. Practically in INS/GPS integration, the system noises are not known correctly. Therefore, an Adaptive Kalman filter is proposed to merge the data of the two systems accurately. The algorithms are tested using the real data of MEMS-IMU (STIM300) and a single frequency NovAtel GPS receiver for land navigations. The integration results indicate a significant improvement in the accuracy of attitude, velocity and position parameters. Moreover, gyro drift which is the main source of errors in INS parameters is significantly reduced.