Abstract: |
Aiming at the problem that the prediction accuracy of the model obtained by curve fitting in the frequency domain identification of small unmanned helicopters is low, this paper proposes a new identification method combining ant colony optimization algorithm and LEVY method. The initial model structure is obtained through the LEVY method, and the parameters in the initial model structure are optimized by using the advantages of the global optimization characteristic of the ant colony algorithm, thereby obtaining the transfer function model of the small unmanned helicopters. The flight data verification shows that the method can fit the frequency response curve well, improve prediction accuracy of the identification model, and more accurately reflect flight dynamic characteristics of the small unmanned helicopters. |