The presence of faults on feedback sensors can have serious effects on the performance of closed-loop systems. This note describes a method of using an auxiliary neural network controller to proviele the required closed-loop dynamic response even when sensorfaults are present. The effectiveness of the scheme is described in relation to a helicopter flight control system which operated with a variety of different types of sensor faults.
Heffley, R.K., Jewell, W.F., Lehman, J.M., and Van Winkle, R.A.1979. A Compilation and Analysis of Helicopter Handling Qualities Data - Vol I. NASA CR-3144.
2.
McLean, D., and Benkhedda, H.1994. 'An Analytical Redundancy Scheme for an Automatic Flight Control System'. In Cook and Rycroft (Eds), Aerospace Vehicle Dynamics and Control, Clarendon Press, Oxford , 145-160.
3.
Patton, R.J., Chen, J., and Nielsen, S.B.1995. 'Model-based methods for fault diagnosis: Some guidelines '. Trans Inst MC, 17(2), 73-83.
4.
Saade, C.1996. 'Neural Control of Hovering Helicopter Flight'. M Phil Thesis, University of Southampton, October.
5.
Zinsch, F.1996. "Application of Neural Networks in Controlling Helicopter Flight'. M Phil Thesis, University of Southampton, October.
