Adaptive Neural Network for a Quadrotor Unmanned Aerial Vehicle

A new adaptive neural control scheme for quadrotor helicopter stabilization at the presence of sinusoidal disturbance is proposed in this paper. The adaptive control classical laws such e-modification presents some limitations in particular when persistent oscillations are presenting in the input. These techniques can create a dilemma between weights drifting and tracking errors. To avoid this problem in adaptive Single Hidden Layer neural network scheme, a new solution is proposed in this work. The main idea is based on the use of two SHL in parallel instead of one in the closed loop in order to estimate the unknown nonlinear function in Quadrotor dynamical model. The learning algorithms of the two SHL Networks are obtained using the Lyapunov stability method. The simulation results are given to highlight the performances of the proposed control scheme.