Fuzzy-neuro fault-tolerant control schemes for aircraft autolanding under actuator failures

In the paper, two fuzzy-neuro control schemes are presented for an aircraft automatic landing problem under the failures of stuck control surfaces and severe winds. The scheme incorporates a fuzzy-neuro controller which augments an existing conventional controller called Baseline Trajectory Following Controller (BTFC). Two fuzzy-neuro controllers have been designed using the recently proposed fuzzy-neuro algorithms named Sequential Adaptive Fuzzy Inference System (SAFIS) and Online Sequential Fuzzy Extreme Learning Machine (OS-Fuzzy-ELM) and a detailed performance comparison has been made. For this study, the following fault scenarios have been considered: i) Single fault of either aileron or elevator stuck at certain deflections and ii) Double fault cases where one aileron and one elevator at the same or opposite direction are stuck at different deflections. The simulation studies indicate that the OS-Fuzzy-ELM achieves better fault-tolerant capabilities compared with SAFIS.