Neural Network‐Based Active Fault‐Tolerant Control for an Eagle‐Inspired Flexible Wing

ABSTRACT This article investigates the active fault‐tolerant control (FTC) problem for a class of eagle‐inspired flexible wings with unknown functions. For the flexible wing, the fault detection (FD) and FTC strategy are combined innovatively. First, the observer‐based FD scheme is established. Since only the system's outputs are assumed to be available, the threshold of FD is only related to the output error. Additionally, neural networks are employed to approximate the unknown functions in the systems. After the fault is detected, the FTC scheme is started. The control scheme can effectively suppress the vibrations of bending deformation and torsional deformation of the flexible wing system caused by faults, and quickly adjust the flight state to ensure that the eagle‐inspired flapping robotic aircraft can still maintain a stable flight attitude and performance in the case of faults. In addition, it is proved that all signals of the closed‐loop system are bounded. Finally, simulation results verify the effectiveness of the proposed method.