Be directed against the characteristics with respect to faster change of aircraft parameters, higher uncertainty and stronger channel coupling, a robust adaptive design method of aircraft attitude controller is proposed. According to the attitude motion equations during unpowered flight, a nonlinear design model facing for attitude control is first established. Then, according to the as-built design model, a robust dynamic inversion control method based on adaptive multi-variable linear extended state observer is proposed, and it is applied to the design of aircraft full-channel attitude controller through dynamic surface control theory. At the same time, a fuzzy parameter adjustment strategy is proposed to adjust the gain coefficient real-timely along with the margin of the present control ability, which can ensure the matching relationship between control parameters and control ability. Through the analysis on the simulation results, in comparison with one conventional controller, the controller proposed in this research ensures that the aircraft can track the attitude angle command faster and more accurately, and it has strong robust characteristics against time-varying uncertainty.