Helicopter nonlinear flight control

Abstract

Due to the inherent instabilities and nonlinearities of rotorcraft dynamics, its changing properties during flight and the difficulties to predict its aerodynamics with high levels of fidelity, helicopter flight control requires strategies that allow to cope with the nonlinearities of the model and assure robustness in the presence of inaccuracies and changes in configuration. The control laws developed in the last years normally concern a complex architecture based on an approximate model inversion, with a robust control synthesis or adaptive elements to compensate for the inversion error. In this thesis, a novel approach based on an incremental model inversion is applied to simplify the design of helicopter flight controllers. With the adopted strategy, by employing the feedback of acceleration measurements to avoid the need for information relative to aerodynamic changes in the rotorcraft, the controller does not need any model data that depends exclusively on the states of the system, thus enhancing its robustness to model uncertainties and disturbances. The control system is composed of a three time scale separated loops architecture that allows to provide navigational control of the vehicle. The overall system is tested by simulating several maneuvers with distinct agility levels commonly used for flying qualities analysis and an efficient tracking of the commanded references is achieved. Furthermore, with the robustness properties verified within the range of inaccuracies expected to be found in reality, the suggested method seems to be eligible for a potential practical implementation, even if only a simplified model of the vehicle is available.