Robust adaptive fault estimation for a commercial aircraft oscillatory fault scenario

A linear time invariant model-based robust fast adaptive fault estimator with unknown input decoupling is proposed to estimate aircraft elevator oscillatory faults. Since the robust fast adaptive fault estimator depends on system output error dynamics which are de-coupled from the unknown inputs (modeling uncertainty), the fault estimation signal generated by the designed fault estimator is robust to the estimated unknown inputs. To obtain a fast fault estimation speed, an adaptive fault estimator involves both proportional and integral components. A Lyapunov stability analysis of the robust fast adaptive fault estimator is given and the fault estimator dynamic response is achieved by pole assignment in subregions realized by LMIs. The proposed robust fast adaptive fault estimator is implemented on a high-fidelity nonlinear aircraft model to detect and estimate elevator actuator oscillatory faults.