Robust Trajectory Tracking Control of Quadrotor UAVs Based on Gaussian Process Learning

A robust control approach based on Gaussian process learning is proposed to solve the problem of trajectory tracking for a quadrotor UAV with parameter uncertainty and wind disturbance. First, Inverse dynamics control (IDC) is used for converting the nonlinear dynamics of the attitude subsystem to a group of double integrators. Then, the upper bound of the model error caused by parameter uncertainty and wind disturbance is estimated by the Gaussian process learning method, and the estimated upper bound is used to compensate the PD controller of the IDC to ensure system stability. The robust control method is also used for position control to achieve high-accuracy tracking. Finally, the simulation results show that the proposed method has better control performance than the MPC-ADRC method.