ERROR ANALYSIS FOR A GUIDED MARS LANDING

This paper presents the methodology and results of a Mars landing error study. Bank maneuvers are used to reduce the landing error due to entry point and atmospheric density errors. The guidance and navigation logic is a derivative of that used for the reentry of the Space Shuttle. Simulations of the guided landing of a biconic vehicle with a maximum L/D of 1.5 show that, using unaided inertial navigation, the approximate bounds on the magnitudes of the downrange and crossrange errors at parachute deployment are 15 km and 1 km, respectively. The limiting error sources are the entry knowledge of downrange and flight path angle and the accelerometer scale factor and bias. The down-range error can be reduced to 5 km if an extended Kalman-filter based navigation logic is used to process the accelerometer data, however, the performance is sensitive to the tuning parameters of the filter. A more robust means of reducing the downrange error is to use ranging relative to either an orbiting vehicle or a vehicle already on the Martian surface, which provides navigation information complementing that derived from the accelerometer data. With range data, the downrange and crossrange errors are both less than 1 km. Similar performance can be achieved with a maximum L/D of 0.5, if the guidance logic is modified so that trajectory corrections are initiated as soon as possible after entry.