导航

ACTA AERONAUTICAET ASTRONAUTICA SINICA ›› 2014, Vol. 35 ›› Issue (2): 532-540.doi: 10.7527/S1000-6893.2013.0341

• Electronics and Control • Previous Articles     Next Articles

Online Aircraft Parameter Identification Using Recursive Fourier Transform

LU Xingju, ZHENG Zhiqiang, GUO Hongwu   

  1. College of Mechatronic Engineering and Automation, National University of Defense Technology, Changsha 410073, China
  • Received:2013-05-23 Revised:2013-07-12 Online:2014-02-25 Published:2013-07-19
  • Supported by:

    National Natural Science Foundation of China (61203095)

Abstract:

Online identification plays a more and more important role in aircraft control system design. Real-time aircraft model updating will allow more intelligent control methods to be adopted. For consideration of velocity and precision of computation, online identification is generally carried on recursively and can be divided into two categories: time-domain methods and frequency-domain methods. Based on the modeling of an aircraft system, the recursive Fourier transform and least square methods are performed in this paper to identify the aerodynamic derivatives and control derivatives online. The approach is validated by simulation of an aircraft's longitudinal model, whose computation and convergence velocities are both fast. The result shows that the identified parameters agree well with the true values, and the algorithm has high adaptability to sensor noise. Finally, the influence of different input signals on identification are investigated, which can serve as reference in real flight identification.

Key words: aircraft, aerodynamic derivative, control derivative, online system, control system identification, Fourier transform, frequency domain, least square

CLC Number: