An Adaptive Cascaded Kalman Filter for Two-antenna GPS/MEMS-IMU Integration
Wei Wang, Zongkai Wu and Hao Zhang, Harbin Engineering University, China
Location: Big Sur
Aiming at the problem of large calculation burden of the traditional nonlinear filtering algorithm and the situation of GPS outages, a novel adaptive cascaded Kalman filtering for two-antenna GPS/MEMS-IMU integration is proposed. The proposed algorithm uses a novel adaptive attitude filter, cascaded with a velocity-position filter. First, the transition relation between additive quaternion error and attitude error is derived, and the attitude filter equation is established. Using the attitude information provided by MEMS-IMU and GPS, the additive quaternion error is calibrated to improve the accuracy of attitude. Then, the velocity-position filter is established to estimate the velocity and position. Cascading the filters, not only the navigation accuracy can be improved , also it can be applied to the situation of GPS outages. The results of driving test show that the proposed algorithm can enhance the precision of attitude, velocity and position estimation, it also effectively overcomes the shortcomings of low attitude precision during GPS outages.