Robust three-stage unscented Kalman filter for Mars entry phase navigation

A high-precision Mars entry navigation algorithm under parameter uncertainties and unknown inputs is developed via a robust three-stage unscented Kalman filter. The uncertainties lie in the atmosphere density, ballistic coefficient and lift-to-drag ratio are the most important error source of state estimation, which have been investigated in this paper. As the robust three-stage unscented Kalman filter has great advantages in parameter estimation and state estimation for nonlinear system with unknown input, it is proposed to compensate the effect of those parameter uncertainties and unknown inputs and obtain more accurate estimation. In the numerical simulation of Mars entry navigation, this filter algorithm has shown its advantages by resulting a high precision navigation with position error less than 100 m and velocity error less than 2 m/s, which fulfills the requirement of future pinpoint Mars landing missions.