Adaptive Neural Control Algorithm Design for Attitude Stabilization of Quadrotor UAV

Abstract

This paper presents a new adaptive neural network control scheme to stabilize the attitude of the quadrotor helicopter. The dynamic model was developed via Newton−Euler formalism. The robust adaptive control is then realized using neural network (NN) algorithm based on a PID controller in the aim to adjust its gain parameters. The proposed algorithm is developed to adapt the structure of the conventional PID controller to a dynamic PID controller. Finally, the proposed controller is compared with that classical PID controller using MATLAB/Simulink. The simulation results show that the neural PID controller produces better performance than the conventional one, particularly in case of perturbation.