Predicting Pilot Opinion Ratings of Flying Qualities of Highly Control-Augmented Aircraft Using an Optimal Pilot Model

摘要

A computer simulation was constructed using optimal control theory to model the human pilot. The purpose of this study was to determine if this optimal pilot model could be used to evaluate highly control-augmented aircraft and predict the human pilot opinion rating of its flying qualities. Eight aircraft/control system configurations were evaluated twice by the model using two different sets of initial conditions. The performance indices for each of the configurations were plotted and compared to a plot of the human pilot ratings of the same configurations. Variances were noted in the relative positioning of the performance ratings and the relative positioning of the pilot opinion ratings, and the mean difference in ratings between the performance index predictions and the pilot ratings varied with the initial conditions of the model (the initial conditions of the actual flight tests were unknown). The RMS error in the state response to system noise for each configuration was also plotted and compared to the pilot ratings. The mean difference in ratings between the RMS error predictions, (which were independent of the model initial conditions), and the pilot ratings compared favorably with the mean difference in ratings between two human pilots, though there appeared to be no correlation between the RMS error and the pilots' comments about the aircrafts' flying characteristics.