摘要
高超声速飞行器因其重要的战略地位已经成为各国争夺空天权所关注的焦点。控制系统设计是保证高超声速飞行器实用化的关键技术。针对吸气式高超声速飞行器,设计一种反步控制器。为增强反步控制器的鲁棒性,引入径向基神经网络对高超声速飞行器纵向非仿射动力学模型中的不确定函数进行在线逼近。为了解决输入受限带来的控制问题,构造一种新型辅助系统对跟踪误差和控制律进行补偿,实现在控制输入瞬时饱和情况下的稳定跟踪。最后,基于MATLAB仿真验证了控制策略的有效性。
Hypersonic vehicles have been becoming the focus in competing for aerospace power because of its important strategic roles.Control system design is the key issue that makes hypersonic vehicles feasible and efficient.Back-stepping control approach is presented for a generic air-breathing hypersonic vehicle(AHV).In order to enhance the robustness of back stepping controller,the radial basis function neural network is applied to approximate the lumped uncertain functions of AHV's longitudinal nonaffine dynamics model.Todeal with input constraints,the novel auxiliary system is developed to compensate both the tracking errors and desired control laws when the input is transiently saturated.Finally,the tracking performance of the proposed control approach is testified based on MATLAB simulations.
作者
王鹏飞
蒋坤
张峰
Wang Pengfei;Jiang Kun;Zhang Feng(Army Academy of Artillery and Air Defence,Hefei 230031,China)
出处
《战术导弹技术》
北大核心
2023年第2期57-65,共9页
Tactical Missile Technology
基金
安徽省青年自然科学基金(2008085QF332)。
关键词
高超声速飞行器
反步控制
输入受限
神经网络
干扰观测器
鲁棒性
hypersonic vehicle
back stepping control
input constraint
neural network
disturbance observer
robustness
