Asynchronous direct Kalman filtering approach for underwater integrated navigation system

This paper presents an asynchronous direct Kalman filter (ADKF) approach for underwater integrated navigation system to improve the performance of the prevalent indirect Kalman filter structure. The designed navigation system is composed of a strap-down inertial navigation system (SDINS) along with Doppler Velocity Log, inclinometer, and depthmeter. In the proposed approach, prediction procedure is placed in the SDINS loop and the correction procedure operates asynchronously out of the SDINS loop. In contrast to the indirect Kalman filter, in the ADKF, the total state such as position, velocity, and orientation are estimated directly within the filter, and further calculations are not performed outside of the filter. This reduces the running time of the computations. To the best of our knowledge, no results have been reported in the literature on the experimental evaluation of a direct Kalman filtering for underwater vehicle navigation. The performance of the designed system is studied using real measurements. The results of the lake test show that the running time of the proposed approach can be improved approximately 7.5% and also the ADKF exhibits an average improvement of almost 20% in position estimate with respect to the prevalent indirect Kalman filter.