Integrated GN&C Analysis

Overview

In the Fall of 2018, I assisted the NASA CCP (Commercial Crew Program) with integrated GN&C analysis for the SpaceX Crew Dragon. The work focused on the oncoming implementation of new GN&C algorithms for the ascent LAS (launch abort system) in early atmospheric flight.

Contributions

I conducted in-depth trade studies for these algorithms and their performance, relative to their predecessors, in NASA’s (Unix-based) 6-DOF Trick simulation environment, and further developed various MATLAB/C++ unit tests and debugged various problems in the simulation source code. Handoff results were used by other GN&C teams and targeted quantifiable understanding of underlying mathematical relations and their influence on the vehicle’s overall trajectory and design parameters including aerodynamic, propulsive, structural, etc.

Timed parachute deployment algorithms were analyzed both in controlled, independent simulations and Monte-Carlo (MC) dispersion analysis. The latter revealed potential hardware failure scenarios in off-nominal cases. Custom Python data processing and analysis tools were also developed for this task, the next task and general use in EG4.

Linear-Optimized Control Allocation algorithms were also analyzed, looking deep into the impact and interrelated dependencies of weighting parameters on flight performance. Numerous weight sets were fine-tuned for a variety of different use-cases. Detailed comparison across the entire flight regime was also made relative to the previous method which used a more traditional control allocation logic.