Feedback Linearization Fractional Order PID Controller Tuned By Neural Networks Theorem

Quadrotor is an underactuated and nonlinear coupled system. in this paper for controlling the dynamic of the system, feedback linearization (FL) method is used to convert the nonlinear model to a simple linear one. Moreover, a combination of fractional order PID (FOPID) with FL is used to improve the tracking ability. Tuning the parameters of NFOPID, because of two more orders of integral and derivation is a complicated task; therefore neural networks (NNs) method is used to cope with this duty. Back propagation (BP) algorithm is used to train the weights of the NNs. Because of the flexibility and online learning of the NNs, the proposed controller can be robust against uncertainties and disturbances. The fractional order operator has infinite dimension and for practical implementation modified Oustaloup's method is used in this paper. and finally the results of the simulations also validate the effectiveness and robustness of the proposed scheme.