Mars Science Laboratory Aerodatabase Trajectory Reconstruction and Uncertainty Assessment

Abstract

In this paper, a method of reconstruction and uncertainty quantification for entry, descent and landing trajectories is applied to the Mars Science Laboratory mission. The method, referred to as the aerodynamic database reconstruction, is a technique founded in previous planetary entry, descent and landing reconstruction analyses. The aerodynamic database algorithm estimates the ambient atmosphere states and wind-relative attitude (angle of attack and angle of sideslip) from the measured aerodynamic forces acting on the vehicle. At every flight condition for which measurements of the aerodynamic forces are available, as recorded by the on-board inertial measurement unit, the measured forces are compared to the forces tabulated in the aerodynamic database. The reconstructed values of atmosphere and wind-relative attitude are those that best reconcile the measured forces with the aerodynamic database. In this manner, state estimates are computed by leveraging data from both flight measurements and pre-flight models. Additionally, uncertainties of the state estimates are computed through an accompanying uncertainty assessment. The uncertainty quantification is an application of a fundamental technique that applies linear covariance propagation to transform input variances into output uncertainties. Flight data from the Mars Science Laboratory entry, descent and landing, having successfully completed on August 5th 2012, is used to reconstruct the in-flight atmosphere and wind-relative attitude of the Mars Science Laboratory trajectory. In order to assess the performance of the aerodynamic database reconstruction, comparisons of the estimated states are made against the extended Kalman filter reconstruction performed by the Mars Entry Descent and Landing Instrumentation reconstruction team. The state uncertainties are evaluated relative to the differences between the two reconstruction approaches in order to assess the accuracy of the estimated states.