Virtual-command-based model reference adaptive control for abrupt structurally damaged aircraft

Abstract Although a high-gain learning rate can offer ideal tracking performance in adaptive control in theory, it can also lead to high-frequency oscillations in practice due to the unmodeled dynamics of the system. In aircraft structural damage scenarios, the strong uncertainty and the safety-critical nature of the problem make this conflict critical. In this paper, a novel virtual-command-based model reference adaptive control (MRAC) scheme for flight control is proposed. In the new framework, the direct relationship between the learning law and the actual tracking error is broken; instead, a virtual command is introduced as the input to the standard MRAC controller. The key feature is that even when the virtual tracking error is large, the actual tracking error can be maintained within a small range; thus, the MRAC learning rate does not necessarily need to be large to suppress the virtual transient tracking error, which is greatly beneficial for the robustness of the MRAC controller. The proposed method is illustrated by the attitude control of the 6-DOF nonlinear Generic Transport Model in a scenario with a broken left wing tip.