Attitude and altitude control of a quadcopter using neural network based direct inverse control scheme

This paper proposes the application of Neural Network based Direct Inverse Control (DIC) for the attitude and altitude control of quadcopter unmanned aerial vehicle (UAV). The backpropagation learning algorithm were utilized in order to find the appropriate connection weights of neurons by using real quadcopter flight data on hovering state. The experimental results showed that the NN-controlled quadcopter can follow the desired trajectory and maintain the hovering state at different levels of altitude with low errors. This results have proven that the performance of the proposed NN DIC controller in controlling a quadcopter UAV is satisfying.