光学 精密工程, 2019, 27 (11): 2392, 网络出版: 2020-01-07   

高超声速飞行器的反步滑模神经网络控制系统

Backstepping sliding mode neural network control system for hypersonic vehicle
作者单位
1 南京航空航天大学 无人机研究院, 江苏 南京 210016
2 中国人民解放军第5311工厂, 江苏 南京 210000
摘要
针对高超声速飞行器一体化气动布局导致弹性机体与推进系统间的强耦合性, 以及跨大空域及高速飞行过程中导致气动特性存在强非线性、不确定性和明显的时变特性, 提出一种基于小脑神经网络的高超声速飞行器反步滑模控制策略。首先建立高超声速飞行器纵向非线性数学模型, 并采用输入-输出反馈线性化方法, 解除多变量之间的耦合关系; 然后设计基于反步法的滑模变结构控制器解决系统非匹配不确定性难题; 同时为弥补反步滑模控制器鲁棒性不足缺点, 利用自回归小脑神经网络(RCMAC)的在线非线性逼近、自学习能力和相应控制结构, 设计基于RCMAC的反步滑模控制器。仿真试验结果表明, 该方法下高超声速飞行器纵向的高度控制精度可达到0.5 m, 速度控制精度为0.1 m/s, 可以保证闭环系统全局稳定, 且拥有良好的跟踪性能和鲁棒性能。
Abstract
Strong coupling between the elastic body and the propulsion system in a hypersonic vehicle is caused by its integrated pneumatic layout and strong nonlinearity uncertainty, and obvious time-varying characteristics of aerodynamics, when the vehicle spans a large airspace and is flying at high speed. To eliminate the influence of this coupling, we propose a backstepping sliding mode control scheme based on a recurrent cerebellar model articulation controller (RCMAC). The input-output feedback linearization approach is used to resolve coupling between multiple variables. Firstly, we established the nonlinear mathematical longitudinal model of a hypersonic vehicle. Secondly, the sliding mode variable structure controller was designed to do away with the uncertainty of mismatch. Finally, the RCMAC-based backstepping sliding mode controller was designed. The controller makes up for the shortcoming of robustness of the hypersonic vehicle by its control structure and ability of nonlinear approximation and self-learning. The results of the simulation experiment indicate that the longitudinal altitude and velocity control precisions of a hypersonic vehicle can reach 0.5 m and 0.1 m/s, respectively and can therefore satisfy the system requirements of global stability, good dynamic responses, and robustness.

刘蓉, 黄大庆, 姜定国. 高超声速飞行器的反步滑模神经网络控制系统[J]. 光学 精密工程, 2019, 27(11): 2392. LIU Rong, HUANG Da-qing, JIANG Ding-guo. Backstepping sliding mode neural network control system for hypersonic vehicle[J]. Optics and Precision Engineering, 2019, 27(11): 2392.

本文已被 1 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!