Adaptive augmenting control (AAC) is a forward gain, multiplicative adaptive algorithm for launch vehicle flight control that meets three summary-level design objectives: Do no harm — return to baseline control design when not needed; respond to errors in the ability of the vehicle to track commands to increase performance; and respond to undesirable parasitic dynamics (e.g., control-structure interaction) to maintain stability.
While classical control methods are typically able to meet requirements for flight, the deployment of advanced control designs has the potential to improve performance capabilities, add robustness, decrease the costs associated with high-fidelity models and tests, and allow recovery from severely off-nominal, unanticipated scenarios.
Even in cases where vehicle damage or degraded performance prevents mission success, adaptive control techniques may be able to delay an abort and/or increase vehicle survivability long enough to improve chances of crew survival, and reduce the risk to property and the public.
The AAC architecture relies heavily on heritage control techniques’ traditional gain-scheduled approach during nominal situations for which the classical design is appropriately tuned. Furthermore, the adaptation limits may be easily mapped back to classical stability margins, the industry standard practice.
AAC is well suited for application on aerodynamically unstable launch vehicles with thrust vector control via augmentation of the baseline attitude/attitude-rate feedback control scheme. It features the ability to not only increase responsiveness in the case of low performance (as with typical adaptive laws), but also to decrease responsiveness in order to regain high-frequency stability. The approach is compatible with standard design features of autopilots for launch vehicles, including phase stabilization of lateral bending and slosh via linear filters.
In addition, the method of assessing flight control stability via classical gain and phase margins is not affected under reasonable assumptions. The algorithm’s ability to recover from certain unstable operating regimes can be understood in terms of frequency-domain criteria. High-fidelity simulation results as well as recent surrogate aircraft flight tests have consistently confirmed AAC’s ability to improve performance and robustness in realistic launch vehicle failure scenarios.