A Self-Tuning Method for Improving the Accuracy of Aircraft Flight Control Simulations
by
J. Mikael Eklund
Queen's University, Kingston, Canada
Coauthors: Michael J. Korenberg

The aircraft simulation industry’s conventional method of creating simulations of aircraft systems is to code in software a mathematical model provided by the aircraft manufacturer on a digital processor, and to interface this model with input and/or output hardware devices used by the pilot. The constraints placed on the software implementation by the processor resources (e.g. iteration rate) and the effects of the hardware devices on the model (e.g. inertial effects) often results in variation between the results obtained in testing the manufacturer’s model alone and when integrated with the complete full flight simulator. This poses a serious problem for the simulator manufacturer which requires many engineering hours to retune the model to compensate for these effects, and often extra engineering and flight testing hours for the aircraft manufacturer who is typically required to provide support and solutions for these problems.

In this paper a method is proposed which will provide a self-tuning method of compensation for some of these effects. This method uses parallel cascade system identification to model the discrepancy between the ideal mathematical model behaviour and the actual simulator behaviour. This method is tested using a detuned flight control model which is adjusted to behave like the correctly tuned model using this method. This method could also be used to adjust the behaviour of an ill-tuned model to match more closely the behaviour of the real system.

Date received: February 10, 1998

Copyright © 1998 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # caav-11.