Design of Neutral Network–Sliding Model Based Large Envelope Flight Control Law

Abstract:

Applying neutral network-sliding model control design methods to large envelope flight control law of aircraft whose model parameter varies greatly with flight condition was studied in this paper. Neural network theory is used to approximately linearize the nonlinear system and cancel the errors brought with approximate inversion, and the residual error is solved by sliding model control. So it can approximate the nonlinear model accurately, and improve robustness and anti-jamming capability of the flight control system. Simulation results show the design neural network – sliding model large envelope flight controller has excellent control performance.

• Periodical: Advanced Materials Research (Volumes 532-533)
• Pages: 503-507
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.532-533.503
• Online since: June 2012
• Authors: Gong Cai Xin, Wei Lun Chen, Jin Niu Tao
• Keywords: Adaptive Control, Neural Network (NN), Nonlinear Control, Sliding Model Control

References

[1] Lewis F L, Yesildirek A, Liu K: Multilayer neural-net robot controller with guaranteed tracking performance [J]. IEEE Transactions on Neural Networks, 7(2): 388-399(1996).

DOI: 10.1109/72.485674

[2] Kim B, Calise A J: Nonlinear flight control using neural networks [J]. Journal of Guidance, Control and Dynamics, 20(1): 26-33(1977).

DOI: 10.2514/2.4029

[3] Calise A J, Rysdyk R: Nonlinear adaptive flight control using neural networks[J]. IEEE Control System Magazine, 18(6): 14-25(1998).

[4] McFarland M B, Calise A J: Adaptive nonlinear control of agile antiair missiles using neural networks[J]. IEEE Trans-actions on Control System Technology, 8(5): 749-756(2000).

DOI: 10.1109/87.865848