Neuro-fuzzy controller for attitude-tracking stabilization of a multi-rotor unmanned aerial system

This paper deals with developing an automatic controller that solves the attitude stabilization for a Quadrotor unmanned aerial system (UAS). The controller used a simultaneous strategy of estimation and compensation of uncertainties as well as disturbances. The approach consisted of integrating a neuro-fuzzy system that implemented a set of differential neural networks (DNNs) as consequence section of Takagi-Sugeno (T-S) fuzzy inference. The combination of these two strategies applied on a Quadrotor UAS has the main purpose of forcing a hover flight while the tracking desired angular positions are attained. The control method identified the unknown nonlinearities and bounded external disturbances firstly. This information served to compensate the uncertain section of the Quadrotor dynamics. An additional section in the controller design enforces the stabilization of the tracking error with respect to a given reference trajectory. The control design methodology supported on the Lyapunov stability theory and guaranteed ultimate boundedness of the identification and tracking errors. Academic simulation tests confirmed the superior performance of the proposed algorithm based on the combination of DNNs and T-S techniques.