Improved pilot training via bifurcation analysis and robust control for aircraft loss of control problems

Abstract

Aircraft loss of control is one of the largest contributors to fatal accidents in the aviation environment. The unprecedented change in aircraft dynamics due to loss of control onset and the associated structural constraints make loss of control prevention and/or recovery a challenging task. State-of-the-art autopilots are generally designed for nominal aircraft operations and disengage under off-nominal conditions, hence cannot be viewed as a safety solution during loss of control onsets. Herein lies the importance of providing training to pilots so as to equip themselves to rescue aircraft from loss of control events. Current pilot training methodologies have significant limitations when it comes to loss of control prevention and recovery strategies. In this context, a simulator for improved pilot training based on bifurcation and continuation techniques is presented in the paper. Augmenting these techniques with the current pilot training procedure can significantly improve the quality of training. This methodology can help pilots to distinguish various loss of control scenarios and aid them in taking appropriate recovery decisions intuitively. Meanwhile, a robust control-based loss of control handling module is also presented for developing non-intuitive strategies for loss of control prevention and recovery. This module can simulate adequate control profiles that the pilot can follow to get in and out of various loss of control scenarios. Moreover, it can be used as pilot activated recovery system in case of pilot disorientation and as a fully autonomous recovery system for much complex scenarios. The simulator is developed in MATLAB/SIMULINK platform and is shown to realize diverse loss of control events like spiral, spin, etc., and subsequent recovery from the same.