Practical Ascpects of Aircraft Flight Control Simulation Using Parallel Cascade Models
by
J. Mikael Eklund
Queen's University, Kingston, Canada
Coauthors: Michael J. Korenberg

In previous work by the authors it was shown that parallel cascade system identification could be used to find models that reproduce the behaviour of an aircraft flight control system. The method is highly efficient in terms of the data collection required for the modelling process since only the input and the output signals of the flight control system are required, and details of the inner workings of the system can be largely ignored resulting in significantly fewer real data signals being needed compared to traditional modelling methods. The parallel cascade model behaviour was compared to the target flight control model behaviour met the criteria used by the U.S. Federal Aviation Authority to certify commercial Full Flight Simulators.

In addition to being able to reproduce the behaviour of the real aircraft system, the new model must be implemented on a full flight simulator. In this paper methods are shown which allow this to be done. Firstly, for a model of the flight control system to used on a full flight simulator it must be capable of running at a high iteration rate (typically 3000 Hz). The memory of the system behaviour can be up to a few seconds in length which can be an excessive computational burden at this iteration rate. A method is used in which the model runs at 3000 Hz but samples the data at a much lower rate. Secondly, the industry standard flight control system uses a hydraulic actuator to drive the pilot flight control (e.g stick or wheel) and these high performance devices typically use PID controllers. Thus any new model would have to provide PID output signals. A method is given which allows for these signals to be computed directly from the parallel cascade models with no additional identification required.

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-10.