Adaptive Kalman Filter-Based Sensor Fault Detection, Isolation, and Accommodation for B-747 Aircraft

Sensor fault detection, isolation, and accommodation via Adaptive Kalman Filter (AKF) algorithm are applied to the lateral dynamics of Boeing-747 aircraft in this study. The flight dynamic model of Boeing-747 aircraft in steady state flight condition is presented and investigated. In nominal case, the Optimal Linear Kalman Filter (OLKF) gives fine estimation values. However, when there is a malfunction on the measurement channels, the accuracy of the filter estimations become poor and the filter becomes unreliable. Two faulty scenarios are investigated. The first scenario comprises the single sensor fault and the second is a simultaneous double sensor fault. The fault detection algorithm detects the fault and isolation process performs via calculating and comparing the statistics of sample and theoretical error variances to distinguish the faulty sensor. Lastly, fault accommodation process is presented in the study as implemented by Adaptive Kalman filter algorithm and demonstrates very efficient, firm, and reliable performance on behalf of enhancing the estimation values of the filter.