Frequency Domain Identification of a Small-Scale Helicopter with Least Square Method

jihong zhu,guanlin wang,jiali yang
DOI: https://doi.org/10.2514/6.2013-4678
2013-01-01
Abstract:Accurate and effective modeling of the helicopter system is an important prerequisite for the implementation of the autonomous flight. This paper makes a modification of the Levy frequency domain least square identification method and uses the modified method to identify the dynamics of a small-scale helicopter near the hover state. We first obtain the real flight data through experimentation and transform the data to frequency domain, then simplify the dynamics of the helicopter to four single input/single output subsystems, and identify each channel's transfer function using the frequency domain least square method, which transforms the identification to the complex curve-fitting problem. The identified models are verified in the time domain with the doublet flight data and the results show that the identification models can accurately and effectively describe the dynamics of the helicopter near the hover point. Moreover, a comparison of the results' validity is made to show the necessity of the Levy method's modification.
What problem does this paper attempt to address?