An Observable Degree-based Adaptive Hybrid Correction Method for Integrated Navigation System

To improve the reliability, accuracy, and anti- interference ability of the navigation system, a combined navigation system of strap-down inertial navigation system/ global navigation satellite system/ polarization skylight sensor/ magnetic field sensor multi-sensor fusion is proposed. The observability based adaptive information fusion method has great potential for improving the high-precision attitude determination and navigation capabilities of integrated navigation systems. For integrated navigation systems with feedback correction, we define a new observable degree using weighted least square method. Since the observable degree of the system state matches the estimation precision to some extent, the normalized observable degree of system state is used during correction. The adaptive hybrid correction of the navigation system can effectively reduce the influence of low-precision estimates. Finally, simulation demonstrate the effectiveness of the proposed adaptive hybrid correction method based on observability analysis.