Reconfigurable flight control system design using adaptive neural networks

被引:86
作者
Shin, DH [1 ]
Kim, Y [1 ]
机构
[1] Seoul Natl Univ, Inst Adv Aerosp Technol, Dept Aerosp Engn, Seoul 151741, South Korea
关键词
adaptive controller; backstepping; neural networks; reconfigurable flight control system;
D O I
10.1109/TCST.2003.821957
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
An adaptive controller design method based on neural network is proposed for reconfigurable flight control systems in the presence of variations in aerodynamic coefficients or control effectiveness deficiencies caused by control surface damage. The neural network based adaptive nonlinear controller is developed by using the backstepping technique for command following of the angle of attack, sideslip angle, and bank angle. On-line learning neural networks are implemented to compensate the control effectiveness decrease and guarantee the robustness to the uncertainties due to aerodynamic coefficients variations. The main feature of the proposed controller is that the adaptive controller is designed by assuming that all of the nonlinear functions of the system have uncertainties, whereas most of the previous works assume that only some of the nonlinear functions are unknown. Neural networks learn through the weight update rules that are derived from the Lyapunov control theory. The closed-loop stability of the error states is also investigated. A nonlinear dynamic model of a high performance aircraft is used to demonstrate the effectiveness of the proposed control law.
引用
收藏
页码:87 / 100
页数:14
相关论文
共 24 条
[1]   NONLINEAR CONTROL LAW WITH APPLICATION TO HIGH ANGLE-OF-ATTACK FLIGHT [J].
BUGAJSKI, DJ ;
ENNS, DF .
JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 1992, 15 (03) :761-767
[2]   Development of a reconfigurable flight control law for tailless aircraft [J].
Calise, AJ ;
Lee, S ;
Sharma, M .
JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 2001, 24 (05) :896-902
[3]  
Ge SS, 1998, P AMER CONTR CONF, P1043, DOI 10.1109/ACC.1998.703568
[4]   NONLINEAR FLIGHT CONTROL DESIGN VIA SLIDING METHODS [J].
HEDRICK, JK ;
GOPALSWAMY, S .
JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 1990, 13 (05) :850-858
[5]   MULTILAYER FEEDFORWARD NETWORKS ARE UNIVERSAL APPROXIMATORS [J].
HORNIK, K ;
STINCHCOMBE, M ;
WHITE, H .
NEURAL NETWORKS, 1989, 2 (05) :359-366
[6]  
IOANNOU PQ, 1996, ROBUST ADAPTIVE CONT, pCH8
[7]   Analysis of soil nailed earth slope by discrete element method [J].
Kim, JS ;
Kim, JY ;
Lee, SR .
COMPUTERS AND GEOTECHNICS, 1997, 20 (01) :1-14
[8]   FLIGHT CONTROL DESIGN USING NON-LINEAR INVERSE DYNAMICS [J].
LANE, SH ;
STENGEL, RF .
AUTOMATICA, 1988, 24 (04) :471-483
[9]   Nonlinear adaptive flight control using backstepping and neural networks controller [J].
Lee, T ;
Kim, Y .
JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 2001, 24 (04) :675-682
[10]   Analysis of adaptive neural networks for helicopter flight control [J].
Leitner, J ;
Calise, A ;
Prasad, JVR .
JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 1997, 20 (05) :972-979