Improved model reference-based adaptive nonlinear dynamic inversion for fault-tolerant flight control

Abstract

Due to lack of robustness, Nonlinear Dynamic Inversion (NDI) control cannot resist the influence of disturbances. In this paper, an adaptive fault-tolerant control method, called Improved Model Reference-based Adaptive NDI (IMR-ANDI), is proposed and applied to develop a fault-tolerant flight controller to improve robustness to faults and disturbances. To improve the estimation accuracy, this paper also modified the piecewise-constant adaptation, which improves the accuracy of adaptive estimation without increasing the sampling frequency. Moreover, a low-pass filter is introduced to the control law creatively to weaken the disturbance of high-frequency dynamics to the adaptation, which balances the dynamic performance and robustness as well as prevents the system oscillation. Finally, a fault-tolerant flight controller is developed based on the IMR-ANDI and verified under the actuator fault and center of gravity variations. All simulation and hardware-in-the-loop experiment results conform to the theoretical analysis, indicating that the developed IMR-ANDI fault-tolerant controller is robust to faults and disturbances and can achieve expected control performance.