Improved BP Network-Based Sliding Mode Tracking Control for a Quadrotor UAV

This paper introduces a novel trajectory tracking control method for quadrotor unmanned aerial vehicles (UAVs). In order to improve the stability of a UAV during trajectory tracking, the proposed method combined the advantages of a back propagation (BP) network and sliding mode control (SMC). A nonlinear dynamic model was first built for a quadrotor UAV by using Newton-Euler equations. The process was then divided into quasi-fully actuated and quasi-under actuated subsystems. The corresponding control law was devised with SMC. For the quasi-under actuated system, the desired roll and pitch angles were calculated according to the desired state, so as to reduce the correction error. For the quasi-fully actuated system, based on the standard BP network, the sliding mode switching function coefficient can be dynamically adjusted following the error by introducing a variable learning rate factor and a momentum factor. The simulation results show that the proposed method could accurately track the desired trajectory, and greatly suppress the chattering phenomenon with the maximum overshoot of the SMC.