Statistical and adaptive approach for verification of a neural-based flight control system

作者: R.L. Broderick

DOI: 10.1109/DASC.2004.1390736

关键词:

摘要: This work presents a combined statistical and adaptive approach for the verification of an adaptive, online learning, sigma-pi neural network that is used aircraft damage flight control. Adaptive control systems must have ability to sense its environment, process dynamics, execute actions. project was completed class in complex at Nova Southeastern University. Verification neural-based system currently urgent significant research engineering topic since these are being looked upon as new survivability, both commercial military applications. The most shortcoming prior current approaches verifying networks application linear non-linear problem. Advances computational power techniques estimating aerodynamic stability derivatives provide opportunity real-time New needed substantially increases confidence use life, safety, mission critical systems.

参考文章(12)
D.S. Phatak, I. Koren, Complete and partial fault tolerance of feedforward neural nets IEEE Transactions on Neural Networks. ,vol. 6, pp. 446- 456 ,(1995) , 10.1109/72.363479
Anthony J Calise, Rolf T Rysdyk, Nonlinear adaptive flight control using neural networks IEEE Control Systems Magazine. ,vol. 18, pp. 14- 25 ,(1998) , 10.1109/37.736008
Joseph Totah, Simulation evaluation of a neural-based flight controller Flight Simulation Technologies Conference. ,(1996) , 10.2514/6.1996-3503
Edward Ott, Chaos in Dynamical Systems Cambridge University Press. ,(2002) , 10.1017/CBO9780511803260
Garnett P. Williams, Chaos Theory Tamed ,(1997)
Manu Sharma, Anthony J Calise, Seungjae Lee, Development of a Reconfigurable Flight Control Law for the X-36 Tailless Fighter Aircraft AIAA Guidance, Navigation, and Control Conference and Exhibit. ,(2000) , 10.2514/6.2000-3940
Edward Ott, Chaos in dynamical systems ,(1993)
Rolf Rysdyk, Anthony Calise, FAULT TOLERANT FLIGHT CONTROL VIA ADAPTIVE NEURAL NETWORK AUGMENTATION Guidance, Navigation, and Control Conference and Exhibit. ,(1998) , 10.2514/6.1998-4483
Richard Durbin, David E. Rumelhart, Product units: a computationally powerful and biologically plausible extension to backpropagation networks Neural Computation. ,vol. 1, pp. 133- 142 ,(1989) , 10.1162/NECO.1989.1.1.133
Gerald E. Peterson, Foundation for neural network verification and validation Proceedings of SPIE. ,vol. 1966, pp. 196- 207 ,(1993) , 10.1117/12.152651