For the auto landing precisely of an Unmanned Aerial Vehicle(UAV), the longitudinal control law for the auto landing of the UVA was designed based on Adaptive Internal Model Control (AIMC) principle. By taking a small wheeled UVA as a working platform, the longitudinal nonlinear model was decoupled and linearized. Then, the ground speed and sink rate were selected as control targets and longitudinal control law was designed based on the AIMC and applied to control system design. The filter parameter was adjusted to improve the dynamic characteristics of the system and the model was identified to enhanced the robustness of the system. The AIMC system was simulated digitally under the conditions of ownwind or headwind in a speed of 6 m/s, and the results show that the landing precision of system is in a scope of 30 m for forward or backward directions. Finally, a hardware test platform was established to verify the simulation results and the hardware-in-loop-simulation (HILS) proves the harmony and integrality of the system.