垂直起降系统的互补滑模与情感神经网络控制
Complementary sliding mode and emotional neural network control for vertical takeoff and landing systems
贾超 1高诗鉴 1高铭 1谷海青1
作者信息
- 1. 天津理工大学 电气工程与自动化学院,天津 300384;天津市复杂系统控制理论及应用重点实验室,天津 300384
- 折叠
摘要
针对具有未知扰动的垂直起降机双闭环控制系统,提出一种基于连续径向基情感神经网络的互补积分滑模控制策略.建立了包含未知扰动项的垂直起降机的数学模型,为系统内外环分别设计滑模控制器和互补积分滑模控制器,基于对系统鲁棒性的考量,利用连续径向基情感神经网络逼近系统未知扰动项,给出神经网络权重的自适应律,形成一套新型的复合控制策略,利用Lyapunov方法分析了系统的稳定性.通过对垂直起降机的仿真实验与实物实验进行分析对比,证明了所设计的控制器具有更快的收敛速度、更高的跟踪精度和更佳的鲁棒性.
Abstract
This paper proposes a complementary integral sliding mode control strategy based on a continuous radial basis function emotional neural network for the double closed-loop control system of vertical takeoff and landing aircraft with unknown disturbances. First, a mathematical model of a VTOL aircraft with unknown disturbance terms is developed. Second, the sliding mode controller and the complementary integral sliding mode controller are intended for the system's inner and outer loops. With consideration of system robustness, the continuous radial basis function emotional neural network is employed to approximate the system's unknown disturbance terms. The adaptive rule of neural network weights is presented to construct a new set of composite control techniques. The Lyapunov approach is used to assess the system's stability. The developed controller achieves faster convergence speed, higher tracking accuracy, and superior robustness after the simulation trials and actual trials of the vertical takeoff and landing aircraft are assessed and compared.
关键词
垂直起降机/互补积分滑模/情感神经网络/偏角控制Key words
vertical takeoff and landing aircraft/complementary integral sliding mode/emotional neural network/deflection angle control引用本文复制引用
基金项目
国家自然科学基金项目(62103298)
天津市自然科学基金项目(18JCYBJC87700)
天津市高校中青年骨干创新人才培养计划()
出版年
2024
国家科技期刊平台
NSTL
万方数据