Velocity Prediction Method of Quadrotor UAV Based on BP Neural Network

This paper proposes a method based on BP neural network to predict the velocity of quadrotor UAV, which can effectively solve the problem of velocity error accumulation overtime during the process of inertial navigation of quadrotor UAV. Through dynamics analysis, it is concluded that the velocity of the quadrotor UAV is related to the attitude, acceleration, angular velocity, and control of the UAV. Based on the analysis of the dynamic, a BP neural network is used to construct a mapping network of the velocity of the quadrotor UAV and the attitude, acceleration, angular velocity, and control amount of the UAV. Through analysis and simulation experiments, it is proved that this method can effectively predict the velocity of quadrotor UAV. It can effectively assist the inertial navigation algorithm and improve the positioning accuracy of the UAV.