中文摘要：

为了改善网络随机时延影响网络遥操作机器人系统稳定运行的问题，采用RBF神经网络对网络实测时延进行预测，取得较好的预测效果．结合Smith预估控制，提山了一种基于网络实测时延的遥操作机器人系统神经网络预测控制新方法，仿真实验结果表明了该方案的有效性和鲁俸性。

英文摘要：

The variable time delay of the network is a key problem which affects the stability of teleoperation. The composing theory of the telerobotics with the test result of the network delay is introduced. The RBF neural network is applied to predict the delay. A new predictive controller based on the RBF neural network and the Smith predictive controller is proposed to eliminate the real delay. Simulation results from MATLAB show that the control scheme is effective and robust.