A Postflight Estimation Method for Guidance Instrumental Error of High-Precision Strap-Down Inertial Navigation System

The post-flight estimation of guidance instrument error is called guidance instrument error separation. It estimates the error coefficient of the airborne inertial navigation system by comparing the airborne inertial data received by the ground with the velocity and position data obtained by external measurement tools such as radar. Since the flight time is short and the error coefficient is not entirely appreciable, the traditional post-flight estimation usually adopts optimization methods such as least squares rather than filter-based methods to avoid convergence problems. The conventional separation method is based on the apparent inertial space navigation velocity or position matching, eliminating gravity’s effect. This treatment avoids the calculation of the partial derivative of the complex gravity model. Still, it makes the measurement error no longer independent, so it is not the best estimate of the error coefficient. With the continuous improvement of the accuracy of the strap-down inertial system, it is difficult for traditional methods to obtain valuable parameter estimation results. This paper proposes a new estimation method that employs a novel mixed calculation coordinate system called ”self-conjugate inertial space.” This coordinate system combines the advantages of the Earth-centered inertial coordinate system, the Earth-centered Earth-fixed coordinate system, and another coordinate system that turns in reverse around the Earth to simplify error analysis. The method considers the influence of external measurement data frequency, external measurement noise, telemetry noise, model linearization error, etc., and reduces the calculation error of the traditional error separation model. Numerical simulation and semi-physical experiments show that the proposed method can significantly improve airborne inertial navigation systems’ error coefficient estimation accuracy under typical conditions.