Abstract: The output error of an inertial measurement unit(IMU) in a fiber optic gyro(FOG) strapdown inertial navigation system is divided into bias, scale factor, misalignment and white noise. A system error model is built, and a Kalman filter is designed fusing high precision information to calibrate the IMU online. This is applied into an FOG strapdown system and tested in a vehicle. The test results indicate that the free inertial navigation performance is greatly improved with the online calibration technique. The error model and the inflight estimation presented in this artical effectively estimate the output error of an IMU, and online calibration is realized, so that inertial system real application precision is improved.

Key words: FOG, error model, Kalman filter, online estimation, online calibration