In this paper, a robust adaptive tracking control scheme is proposed for the lateral attitude dynamic of an aircraft in the presence of unmeasured unsteady aerodynamic disturbance, external disturbance, system uncertainty, and input saturation at high angle of attack (AOA). To improve the robust control performance, the unknown system uncertainty is estimated using the adaptive radial basis function neural network (RBFNN). Meanwhile, the nonlinear disturbance observer (NDO) is developed to estimate the compounded disturbance, which consists of the unmeasured unsteady aerodynamic disturbance, the external disturbance and the unknown neural network approximation error. Furthermore, considering the actuator saturation problem, an auxiliary system is designed to weaken the input saturation. Then, based on the comprehensive design of the RBFNN, the NDO and the auxiliary system, a robust adaptive attitude tracking control law for the lateral attitude dynamic of an aircraft is developed. Meanwhile, the dynamic surface technique is employed to avoid the tedious differential computations for the virtual control law. The strict Lyapunov analysis proves that the signals of the closed-loop system are bounded. Finally, simulation results are shown to illustrate the effectiveness of the proposed robust adaptive attitude tracking control scheme.