摘要
针对倾转翼飞机过渡段控制存在的时变、欠驱动、强耦合等非线性特点,采用滑模控制来对其进行控制,然后在此基础上引入RBF神经网络,利用其非线性映射能力有效解决了滑模控制中存在的误差问题,进一步改善了系统的动态性能。研究表明,采用基于RBF神经网络的滑模控制方法,可有效提高倾转翼飞机过渡段定高飞行的控制精度,同时也证明了在处理时变、欠驱动、强耦合的非线性系统时,滑模控制与神经网络结合具有其独特的优势。
Aiming at the time-varying,under-actuated,strong coupling and other nonlinear characteristics of tilt-wing aircraft transition control,sliding mode control was used to control the tilt-wing aircraft transition section.Then RBF neural network was introduced to solve the error problem of sliding mode control effectively by using its nonlinear mapping ability.The dynamic performance of the system was improved.The results show that the sliding mode control method based on RBF neural network can effectively improve the control accuracy of the tilting wing aircraft in the transition phase.It also proves that the combination of sliding mode control and neural network has its unique advantages in dealing with time-varying,under-actuated and strongly coupled nonlinear systems.
作者
吴健健
王琦
李之瀚
刘阳
WU Jian-jian;WANG Qi;LIU Yang;LI Zhi-han(Nanchang Hangkong University,Nanchang Jiangxi 330063,China)
出处
《计算机仿真》
北大核心
2020年第4期48-51,共4页
Computer Simulation
基金
江西省科技厅重点研发计划项目(20151BBE50026)。
关键词
倾转翼
耦合
神经网络
滑模控制
Tilting wing aircraft
Coupling
RBF neural network
Sliding mode control