Assessment of helicopter model fidelity through inverse simulation

The paper presents a technique for assessing the reliability of a set of helicopter models in predicting the required control action when executing a given (set of) manoeuvre task(s). An inverse simulation algorithm based on the integration method is used in order to derive the time-history of control commands necessary for following a prescribed ight path. A quantitative comparison between the control laws thus obtained is performed in order to assess the reliability of lower order models with respect to the baseline, most complete one, adopted as a reference for the analysis. Two metrics are developed, one for evaluating a global error level in the definition of the required control law, and a second one for the identification of the uncertainty in the control action when adopting a lower order model. A total of 9 main rotor dynamic models, 3 main rotor inow models and 3 fuselage aerodynamic databases are combined in order to obtain as many as 13 difierent helicopter simulation models, analyzed in 3 manoeuvres: a hurdle-hop, a slalom and a lateral repositioning. The evaluation of the uncertainty associated with the command law identified by means of simpler models is thus performed in terms of the considered metrics, the validity of which is then tested on two more manoeuvres: a pop-up-pop-down manoeuvre and a 180 deg fast turn. The results show that most of the times uncertainty intervals are correctly identified, although with some degree of conservativeness, when less demanding manoeuvres are dealt with. © 2011 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.