Aiming at the need for high-precision error compensation, higher application accuracy and faster startup speed of the fiber optic platform system, the article takes the error characteristics such as the lever arm error of the accelerometer, time asynchronous error of the inertial instruments, the installation error and the scale factor error into account, the two-level calibration Kalman filter equation is established. The Kalman filtering method is used to identify the error parameters, and the method is simulated and verified by experiments. The simulation and experimental results show that the designed system-level calibration algorithm can estimate all error parameters and has higher application performance, which is of great significance to improve the application accuracy of the high-precision fiber optic platform system.