当前位置: X-MOL 学术Proc. Inst. Mech. Eng. Part G J. Aerosp. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Disturbance observer–based neural flight control for aircraft with switched time-varying distributed delays
Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering ( IF 1.0 ) Pub Date : 2021-03-26 , DOI: 10.1177/0954410021998501
Dawei Wu 1 , Jun Zhou 1 , Hui Ye 2
Affiliation  

In this article, the high angle of attack (AOA) maneuver control problem is studied under multiple disturbances and uncertainties. For the first time, the switched distributed delay is constructed to characterize the unsteady aerodynamics. Based on neural networks (NNs) and hyperbolic tangent function, the disturbance observer technique is extended to the nonstrict-feedback system control. To handle the switching problem, time-delay problem, and nonstrict-feedback problem caused by switched distributed delay terms, the Lyapunov–Krasovskii (LK) functional method and a variable separation method are cleverly combined. The proposed LK function can relax the constraints on time-varying delay. Finally, a disturbance observer–based neural finite-time prescribed performance flight control law is developed to improve the flight performance at high AOA, and its effectiveness has been verified through rigorous theoretical analysis and simulation experiments.



中文翻译:

带有时变分布时滞的基于扰动观测器的神经飞行控制

本文研究了多种扰动和不确定性条件下的高攻角机动控制问题。开关延迟分布首次被构造来表征不稳定的空气动力学特性。基于神经网络(NNs)和双曲正切函数,将扰动观测器技术扩展到非严格反馈系统控制。为了处理由切换的分布式延迟项引起的切换问题,时滞问题和非严格反馈问题,巧妙地结合了Lyapunov–Krasovskii(LK)函数方法和变量分离方法。所提出的LK函数可以放宽对时变延迟的约束。最后,开发了一种基于干扰观测器的神经有限时间规定性能的飞行控制律,以改善高AOA时的飞行性能,

更新日期:2021-03-27
down
wechat
bug