Model Reference Adaptive Control of a Quadrotor UAV based on RBF Neural Networks

Abstract

In this paper, a model reference control method based on RBF neural networks is applied to attitude control of quadrotor. The model of a quadrotor is constructed and simplified to obtain the reference model in the same order as the plant. The RBF is trained by using the gradient descent method. Through simulation experiments, MRAC based on RBF has presented good tracking performance on the nonlinear quadrotor system with unknown and changing parameters.