Sage Journals: Discover world-class research

Abstract

The use of neural network controllers to obtain structural load alleviation of a large transport aircraft which exhibits a significant number of bending modes is novel in aircraft flight control systems. In this paper some details of a mathematical model of the aircraft dynamics which was used in the development of the controllers is presented first. Next, details relating to a linear, continuous optimal LQR controller (the baseline controller) are presented. The results obtained from a digital simulation of the optimal structural load alleviation control system (SLACS) then follow. These results and many others obtained from the same simulation were used as the training data required by the neural network controllers used in the design. Two different types of neural network were considered for use, and information about the corresponding training performance and control effectiveness is also presented. The dynamic performance achieved by using the preferred neural controller is compared with that obtained using the baseline LQR controller. A number of suggestions are made in the conclusions in respect of the choice of type of neural network for use as controllers in a SLACS.

Keywords

Active control technology optimal control LQR load alleviation neural networks aircraft flight control

Get full access to this article

View all access options for this article.

References

Ahmed-Zaid, F. , Ioannou, P.A. , Polycarpou, M.M. , and Youssef, H.M. 1993. 'Identification and control of aircraft dynamics using radial basis function, Proc 2nd IEEE Conf Cont Appl, Vancouver, BC, Canada, 567-572.

Ali, Q.R. 1987. 'Aircraft load alleviation by specifying its closed-loop eigenstructure', PhD thesis, Loughborough University, England.

Bisplinghoff, R.L. , Ashley, H. , and Halfman, R.L. ,. 1955. Aercoelasticity. Addison Wesley , Reading, USA.

Brown, M. , Harris, C.J. , and Parks, P.C. 1993. 'The interpretation capabilities of the binary CMAC', Neural Networks, 6(3), 429-460.

Bryson, A.E. , and Ho, Y.C. 1975. Applied Optimal Control. J. Wiley and Sons, New York.

Chalk, C.R. , et al. 1969. 'Background Information and User's Guide for MIL-8785B(ASD): Military Specification - Flying Qualities of Piloted Aircraft ', AFFDL-TR-69-72, August.

Fahlman, S.E. 1988. 'An Empirical Study of Learning Speed in Back-Propagation Networks', CMU Technical Report, CMU-CS-88-162, June.

Harvey, C.A. , and Pope, R.E. 1977. 'Study of the synthesis techniques for insensitive aircraft control systems', Nasa CR 2803, April.

Johnson, S.J. 1978. 'Aircraft Model Prototypes which have Specified Handling Qualities Time Histories', NASA CR-142848.

10.

Jorgenson, C.C. , and Schley, C. 1990. 'A neural network baseline problem for control of aircraft flare and touchdown'. In: Miller, W. T. , Sutton, R. S. and Werbos, R. S. (Eds), Neural Network for Control , MCT Press, Mass, USA, 401-425.

11.

Lapedes, A. , and Farber, R. 1987. 'Non-linear Signal Processing using Neural Networks: Prediction and System Modelling, Los Alamos Report, LA-UR-87-2662.

12.

McLean, D. 1990. Automatic Flight Control Systems, Prentice Hall, Hemel Hempstead, UK.

13.

McLean, D. , and Prasad, R.A. 1980. 'A structural load alleviation control system for a large transport aircraft', TT 8002, July, Loughborough University , UK.

14.

McLean, D. 1983. 'Analytically Redundant Output Feedback Scheme for Reduction of Structural Loads for a Flexible Transport Aircraft', TT 8303, April, Loughborough University, England.

15.

Milne, W.E. 1964. 'Dynamics of deformable aircraft', ARC R & M 3345, London, UK.

16.

Samad, T. 1989. 'Back-Propagation Extensians', Honey-well SSDC Technical Report, Golden Valley, MN 55427, USA.

17.

Schwanz, R.C. 1972. 'Formulation of the equations of motion of an aeroelastic aircraft for stability and control and flight control applications'. AFFDL/ FGC-TM-72-14, August, Dayton, USA.

Neural control of an aircraft load alleviation control system

Abstract

Keywords

Get full access to this article

References