Simplified 5th-Degree Multiple Fading CKF for GPS/INS Integrated Navigation System
Hang Lu, ShunYi Hao, Zhi-ying Peng · 2018
In view of the problem that the system model mismatch and the state abrupt filtering precision decline in the traditional volumetric Kalman filter (CKF), the strong tracking filter (STF) is combined with the fifth-degree Cubature Kalman filter (CKF5). A Simplified fifth-degree Multiple fading CKF algorithm (SMCKF5) is proposed, which has higher filtering accuracy than traditional CKF, and simplifies the calculation steps of FCKF by using the characteristics of the filtering model. At the same time, multiple fading factors are introduced into CKF5 to enhance the adaptability of the algorithm and the ability to deal with the state mutation. The proposed algorithm is applied to GPS/INS integrated navigation system for simulation experiments. The results show that the real value of the abrupt state can be accurately estimated by the proposed algorithm. The filtering performance is better than that of CKF5, and the self-adaptability and positioning accuracy of integrated navigation system are improved.