Self-calibration rank filter for unknown dynamic inputs mitigation during the Mars powered descent phase

Advanced navigation systems for pinpoint landing are required in the entry, descent and landing phase of future missions to Mars. To overcome the horizontal position estimation problem in the Mars powered descent phase, the inertial measurement unit, Doppler radar and surface beacon integrated navigation scheme is proposed, and a conventional filter such as an extended or unscented Kalman filter is adopted. However, in engineering practice, these conventional filters for nonlinear systems with unknown dynamic inputs may degrade or even diverge. The lack of navigation accuracy may result in a large growth of spurious navigations. To solve this problem, based on the rank filter, a self-calibration rank filter is proposed for state estimation of a nonlinear system to mitigate the effects of unknown dynamic inputs. Monte Carlo simulation results are presented to demonstrate the good performance of the self-calibration rank filter for the Mars powered descent navigation. The self-calibration rank filter not only prevents the divergence of the filtering but also significantly improves the state estimation accuracy.