The Trajectory Control of Quadrotor UAV Based on Backstepping With Adaptive RBF Neural Network

This paper utilize the approximation properties of radial basis function (RBF) neural networks, fast adaptive command filtering for trajectory control of quadrotor UAV under dynamic uncertainty. First, the quadrotor UAV dynamics system is decoupled into two inner- and outer-loop subsystems: the outer-loop position subsystem and the inner-loop attitude subsystem. Secondly, a position controller is designed to realize the tracking of the desired position of the quadrotor UAV using backstepping control method in the position subsystem, and provide the desired desired roll and pitch angles for the attitude subsystem. The RBF neural network adaptive approach is used in the attitude subsystem to design the attitude tracking controller to compensate for the uncertainty of the model disturbances, the computational explosion problem is solved using the command filtering technique. The stability of the closed-loop system is ensured by progressively stabilizing the subsystems and the boundedness of all variables is verified. Simulation results show that the designed controller is able to achieve accurate tracking of the desired flight command pairs.