Adaptive neural control based on HGO for hypersonic flight vehicles

作     者：XU Bin 1,2 , GAO DaoXiang 3 & WANG ShiXing 1 1 Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China 2 Autonomous System Lab, ETH Zurich Zurich 8092, Switzerland 3 School of Technology, Beijing Forestry University, Beijing 100083, China 

作者机构：Department of Computer Science and Technology Tsinghua University Autonomous System Lab ETH Zurich School of Technology Beijing Forestry University 

出 版 物：《Science China(Information Sciences)》 (中国科学:信息科学(英文版))

年 卷 期：2011年第54卷第3期

页      面：511-520页

核心收录：

学科分类：0810[工学-信息与通信工程] 0808[工学-电气工程] 08[工学] 081105[工学-导航、制导与控制] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by Sino Swiss Science and Technology Cooperation, the National Natural Science Foundation of China (Grant Nos. 60625304, 90716021) the National Basic Research Progran of China(Grant Nos. G2007CB311003, 2009CB724002) 

主　　题：adaptive neural control hypersonic flight vehicle high gain observer output-feedback 

摘      要：This paper describes the design of adaptive neural controller for the longitudinal dynamics of a generic hypersonic flight vehicle (HFV) which are decomposed into two functional systems, namely the altitude subsystem and the velocity subsystem. For each subsystem, one adaptive neural controller is investigated based on the normal output-feedback formulation. For the altitude subsystem, the high gain observer (HGO) is taken to estimate the unknown newly defined states. Only one neural network (NN) is employed to approximate the lumped uncertain system nonlinearity during the controller design which is considerably simpler than the ones based on back-stepping scheme with the strict-feedback form. The Lyapunov stability of the NN weights and filtered tracking error are guaranteed in the semiglobal sense. Numerical simulation study of step response demonstrates the effectiveness of the proposed strategy in spite of system uncertainty.