In this study, we designed a terrain following framework using model predictive control and nonlinear disturbance observer-based backstepping for a tiltrotor unmanned aerial vehicle (UAV). Terrain following with pitch hold is important in the operation of a tiltrotor UAV because of the uncooled infrared (IR) system and the avoidance of detection. Furthermore, a small tiltrotor UAV is more sensitive to wind disturbance near mountains. Therefore, we propose a nonlinear disturbance-based backstepping method for the flight path angle to enhance flight performance. Model predictive control is used to obtain the optimal flight path angle considering the constraints on the pitch and altitude. The numerical simulation results were compared with the model predictive control backstepping to verify the performance of the proposed framework.