Adaptive recurrent-functional-link-network control for hypersonic vehicles with atmospheric disturbances  

作　　者：DU YanLi WU QingXian JIANG ChangSheng XUE YaLi 

机构地区：[1]College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China [2]College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China

出　　处：《Science China(Information Sciences)》2011年第3期482-497,共16页中国科学（信息科学）（英文版）

基　　金：supported by the National Natural Science Foundation of China (Grant Nos. 90716028, 60974106)

摘　　要：The controller design for a near-space hypersonic vehicle （NHV） is challenging due to its plant uncertainties and sensitivity to atmospheric disturbances such as gusts and turbulence. This paper first derives 12 states equations of NHVs subjected to variable wind field, and presents a novel recurrent neural network （RNN） control method for restraining atmospheric disturbances. The method devises a new B-spline recurrent functional link network （BRFLN） and combines it with the nonlinear generalized predictive control （NGPC） algorithm. Moreover, the proportional-derivative （PD） correction BRFLN is proposed to approximate atmo- spheric disturbances in flight. The weights of BRFLN are online tuned by the adaptive law based on Lyapunov stability theorem. Finally, simulation results show a satisfactory performance for the attitude tracking of the NHV in the mesosphere, and also illustrate the controller＇s robustness to wind turbulence.The controller design for a near-space hypersonic vehicle （NHV） is challenging due to its plant uncertainties and sensitivity to atmospheric disturbances such as gusts and turbulence. This paper first derives 12 states equations of NHVs subjected to variable wind field, and presents a novel recurrent neural network （RNN） control method for restraining atmospheric disturbances. The method devises a new B-spline recurrent functional link network （BRFLN） and combines it with the nonlinear generalized predictive control （NGPC） algorithm. Moreover, the proportional-derivative （PD） correction BRFLN is proposed to approximate atmo- spheric disturbances in flight. The weights of BRFLN are online tuned by the adaptive law based on Lyapunov stability theorem. Finally, simulation results show a satisfactory performance for the attitude tracking of the NHV in the mesosphere, and also illustrate the controller＇s robustness to wind turbulence.

关 键 词：near-space hypersonic vehicle atmospheric disturbances nonlinear predictive control ADAPTIVECONTROL B-spline recurrent functional link network 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] V43[自动化与计算机技术—控制科学与工程]