摘要
采用2个对角递归神经网络(DRNN)构成非线性逆动态控制系统,一个用作辨识器,逼近系统的正模型,为逆动态控制提供系统的灵敏度;另一个用作控制器,逼近系统的逆动态模型,再与原系统串联组成伪线性系统.仿真结果表明该控制方法实现了对非线性系统的有效控制.
Two diagonal recurrent neural networks (DRNN) are utilized to realize the inverse dynamic control strategies for nonlinear systems. One is used as an identifier to approximate the positive model which provides the sensitivity information of the system, and the other is used as a controller to approximate the inverse dynamic model and then it is combined with the system to construct a pseudo-linear system. The simulation results have shown that the method has a much better performance in the non-linear systems.
出处
《甘肃科学学报》
2005年第2期31-33,共3页
Journal of Gansu Sciences
关键词
逆动态控制
对角递归神经网络
反馈线性化
非线性系统
inverse dynamic control
diagonal recurrent neural networks (DRNN)
feedback lineariza-tion
nonlinear system
