航空学报 > 2025, Vol. 46 Issue (17): 331654-331654   doi: 10.7527/S1000-6893.2025.31654

基于在线辨识的高速变构飞行器强适应控制

刘泓麟1, 王冠1, 安帅斌1, 马少捷2, 刘凯1,3()   

  1. 1.大连理工大学 力学与航空航天学院,大连 116024
    2.中国运载火箭技术研究院 研究发展部,北京 100076
    3.大连理工大学 工业装备结构分析优化与CAE软件全国重点实验室,大连 116024
  • 收稿日期:2024-12-11 修回日期:2025-04-07 接受日期:2025-05-08 出版日期:2025-06-03 发布日期:2025-05-27
  • 通讯作者: 刘凯 E-mail:carsonliu@dlut.edu.cn
  • 基金资助:
    基础科研计划(JCKY2022110C019);教育部联合基金项目(8091B032223);中央高校基本科研业务费资助(DUT24RC(3)067)

Online identification based strong adaptive control of hypersonic morphing vehicles

Honglin LIU1, Guan WANG1, Shuaibin AN1, Shaojie MA2, Kai LIU1,3()   

  1. 1.School of Mechanics and Aerospace Engineering,Dalian University of Technology,Dalian 116024,China
    2.Research and Development Center,China Academy of Launch Vehicle Technology,Beijing 100076,China
    3.Key Laboratory of Structural Analysis,Optimization and CAE Software for Industrial Equipment,Dalian University of Technology,Dalian 116024,China
  • Received:2024-12-11 Revised:2025-04-07 Accepted:2025-05-08 Online:2025-06-03 Published:2025-05-27
  • Contact: Kai LIU E-mail:carsonliu@dlut.edu.cn
  • Supported by:
    Industrial Technology Development Program(JCKY2022110C019);Joint Fund Project of the Ministry of Education(8091B032223);Fundamental Research Funds for the Central Universities(DUT24RC(3)067)

摘要:

针对高速变构飞行器在模型不确定、外界扰动以及非最小相位特性影响下的强适应控制问题,首先构建包含不确定性的动力学模型,依据模型特性将其分解为速度系统和姿态系统。设计基于修正遗忘因子的在线参数辨识方法,实时获取本体气动参数,降低对模型知识的依赖程度,为控制器实时提供模态评估信息。然后,提出博弈增强神经网络观测器处理包含辨识误差、变构不确定性、外界扰动在内的复合扰动,使系统对干扰的估计误差能在有限时间内收敛至原点。通过设计变构型-重定义策略重定义系统输出,针对性地给出变构型下的重定义参考指令,避免由升降舵和升力耦合产生的不稳定内动态。最后基于Lyapunov理论进行了系统稳定性分析,通过仿真验证了所提出方法的有效性。

关键词: 高速变构飞行器, 非最小相位, 强适应控制, 在线辨识, 扰动观测

Abstract:

This paper investigates the robust adaptive control problem for high-speed morphing aircraft under model uncertainties, external disturbances, and non-minimum phase characteristics. First, a dynamic model incorporating uncertainties is established and subsequently decomposed into velocity and attitude subsystems based on system characteristics. A modified forgetting factor-based online parameter identification method is designed to estimate aerodynamic parameters in real-time, reducing reliance on prior model knowledge while providing real-time mode evaluation information for controller design. Subsequently, a game-enhanced neural network observer is proposed to handle composite disturbances, including identification errors, morphing uncertainties, and external disturbances, ensuring finite-time convergence of disturbance estimation errors to zero. By developing a morphing-redefinition strategy to reconfigure system outputs, specifically defined reference commands are generated for morphing configurations to avoid unstable internal dynamics caused by the coupling between elevators and lift. Finally, system stability is rigorously analyzed using Lyapunov theory, with simulation results validating the effectiveness of the proposed methodology.

Key words: hypersonic morphing vehicles, non-minimum phase, strong adaptive control, online identification, disturbance rejection

中图分类号: