We applied dynamic neural network algorithm to the controller and the err path. This paper provides multi - channel dynamic neural network algorithm ( MDNN ). 文中提出了多通道动态神经网络(MDNN)算法,该算法把控制器算法和抵消路径的辨识结合在一起考虑,并都用动态神经网络来实现。
Using dynamic recurrent neural networks as identification and controller, the minimum error control of robot tracking the idea locus is implemented. 采用动态对角回归神经网络作为辨识器和控制器,实现了机器人轨迹跟踪的最小误差控制。
To remedy this problem, a real-time hysteretic compensation control strategy combining a dynamic recurrent neural network ( DRNN ) feedforward controller and a proportional derivative ( PD ) feedback controller was proposed to implement the precision position tracking control of the GMA. 为了克服这个问题,将动态递归神经网络(DRNN)前馈和PD反馈控制器相结合,提出了一种实时滞回补偿控制策略,以期实现GMA的精密位移跟踪控制。
This method combines PD dynamic feedback controller with fuzzy neural identifier and neural controller, in order to approach the nonlinear dynamics and reduce the tracking error caused by the robotic nonlinear uncertainties. 该方法在PD动态反馈控制的基础上,引入模糊回归神经网络辨识器(FRNNI)和控制器(FRNNC)在线动态逼近对象的非线性动力学,减少了动力学非线性造成的误差。