A Dual-State Filter for a Relative Velocity Aiding Strapdown Inertial Navigation System

Abstract

The strapdown inertial navigation system that uses the aiding velocities from a relative log is usually affected by the ocean current. Therefore, a dual-state filter algorithm to alleviate the effect from unknown ocean current is proposed in this article. Generally speaking, the ocean-current velocity is relatively stable, in other words, two adjacent samples over a short period can be considered to be unchanged. Consequently, our algorithm chooses the difference between the two adjacent sample points from the log as the external measurement information of the integrated navigation system, where the effect from ocean-current velocity will be alleviated in the filter. Nevertheless, the previous state (the previous velocity error) will exist in the filter updating process. The common Kalman filter algorithm cannot cope with this condition, so we rederived filter equations regarding the dual-state filter based on the previous state and the current state. The final results of simulation and experiment demonstrate that the dual-state filter algorithm can alleviate the effect from unknown ocean current.