Robust neural decentralized control for a quadrotor UAV

This paper presents a nonlinear control of a quadrotor unmanned aerial vehicle(UAV) for trajectory tracking. The dynamic model is obtained by the Euler- Lagrange methodology. In this paper, the control strategy of the quadrotor is based on inner (attitude control) and outer (position control) loops. The outer loop generates the inputs for inverse dynamics and calculates instantaneous desired angles for inner loop to stabilize the orientation of the vehicle. In addition, a recurrent high order neural network (RHONN) trained with extend Kalman filter is used to identify motors dynamics, and the inverse optimal control is employed for stabilization of propellers speed. Simulations results present the feasibility of the proposed control strategy. The feasibility of the control scheme is presented in the presence of parameters uncertainties to corroborate the robustness of the proposed strategy.