Accuracy–simplicity trade-off for small-scale helicopter models: A comparative study based on flight data. (April 2018)

• Record Type: Journal Article
• Authors: Roussel, Emmanuel; Gassmann, Vincent; Laroche, Edouard
• Is Part Of: Control engineering practice. Volume 73(2018)
• Page Start: 56
• Page End: 65
• Publication Date: 2018-04
• Subjects: Unmanned aerial vehicle -- Helicopter -- Experimental model validation -- Modeling -- System identification -- Parameter estimation; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2018.01.001
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: A good model is a trade-off between simplicity of the model structure and accuracy of the prediction. Higher accuracy is generally expected from more complex models, but at the cost of higher computational burden, more complex handling and potential identifiability issues. Depending on the targeted use, assumptions and simplifications are made to find the simplest model still capturing the important phenomena. These choices are not straightforward and, to this end, the paper gives a comparison of miniature helicopter models often found in the literature. The contribution of the paper is thus twofold. A time-domain identification procedure for parametric models of miniature helicopters is first described and applied to four different models with increasing complexity. The procedure is based on flight data obtained during a manual slow-speed flight. Secondly, the accuracies of these models are evaluated and compared, which highlights the main differences and improvements brought by the differences in the aerodynamic model equations and allows the selection of a relevant model structure depending on the target application.