In order to improve the flight stability, an anti-disturbance control algorithm for UAV based on multi-dimensional spectral peak combined weighted search is proposed. The spatial dynamics model of UAV flight is constructed in the velocity coordinate system, the body coordinate system and the ballistic coordinate system. The distributed source modeling method is used to analyze the beam directivity characteristics of UAV flight. The steady-state tracking error of UAV is controlled by the weighting coefficient of beamspace direction, and the small disturbance suppression is carried out by combining the width of the main lobe of the beam, and the joint estimation of UAV flight control parameters is realized by using the multi-dimensional spectral peak joint weighted search method. Error feedback and disturbance suppression are carried out according to parameter estimation results to improve flight stability control performance. The simulation results show that the proposed method has better anti-disturbance performance, higher accuracy of joint estimation of flight control parameters and better output stability.