ESA GNC Conference Papers Repository

This paper describes the development and implementation of the Adaptive Augmenting Control (AAC) system for NASA’s Space Launch System, leading to its groundbreaking flight demonstration during the Artemis I launch on November 16, 2022. Particular attention is given to principles that may be useful to flight controls practitioners of future space transportation systems, with an emphasis on the studies, perspectives, and principles that shaped the early concepts and algorithm design refinements. Principles that drove the design architecture included ensuring the adaptive system was structured such that it would improve the baseline flight control system robustness, yet not unacceptably increase risk when the vehicle was operated within its nominal design envelope. Other key principles included maintaining a classical design architecture and relying on the adaptive system only to augment the control commands, an ability to both increase performance (error regulation) and decrease performance to avoid exciting parasitic dynamics such as vehicle flexibility and slosh, and a prescribed bound on the adaptive gains that correlated to classical stability margins. As the design matured, the designers sought the simplest adaptive control law architecture that met the design objectives, multiple stability analysis approaches were investigated, and the architecture was rigorously reviewed and evaluated using both high-fidelity simulations and a surrogate aircraft flight test campaign. The paper will document the evolution of the architecture, discuss analysis techniques employed, and provide an overview of the path that led to a successful transition of the technology from concept to flight. The original design that was tested and developed using a reduced order simulations is shown to evolve systematically to the final architecture that was flown on Space Launch System, the launch vehicle for the Artemis I mission. Analysis techniques employed throughout the AAC design, development, and verification phases will be discussed; these include standard Monte Carlo analyses (comparisons with and without AAC), expanded Monte Carlo analyses that included worst-on-worst dispersions, stressing cases, frequency-domain stability analysis accounting for the effects of AAC, time-domain stability margin assessments, and generalized gain margins based on the circle criterion. Key components of the AAC Technology Readiness Level (TRL) advancement will be described, including Software-in-the-Loop testing in the Marshall Space Flight Center’s System Integration Laboratory (SIL) and a suite of carefully designed flight tests on an F/A-18. In its final test during the uncrewed first flight of SLS, AAC performed as intended throughout the launch phase and its response reflected a flight trajectory within the pre-flight expectations.