Self-calibration scheme of RIMU based on AEKF

In the navigation process, RIMU can work normally, when one of the sensors is broken. When the navigation process finished, in order to reduce costs, only the faulty sensor needs to be changed. Accuracy of RIMU will be reduced, due to the uncalibration sensor. To solve this problem, a self-calibration scheme for RIMU based on adaptive extended Kalman filter (AEKF) is proposed. Firstly, coordinate systems and their relations are given. Then, system model, dynamic model and observation model are derived. Finally, AEKF is proposed to estimate bias, scale errors and align errors. In the proposed scheme, the new sensor can be calibrated accurately without external references. In addition, the proposed scheme shows an excellent adaptability to noise changes.