摘要
针对工业机器人控制系统在实际工作中不可避免地要受到随机噪声的影响,提出了基于卡尔曼滤波的RBF神经网络与PD复合控制器设计,该控制器由PD反馈、RBF神经网络前馈控制器和卡尔曼滤波器三部分构成;基于卡尔曼滤波的PD和RBF神经网络复合控制具有优良的控制性能和动态品质,能快速跟踪设定的参考信号,而且无大的超调和振荡出现,只需较少的控制能量,明显抑制白噪声的污染,提高系统自适应性和鲁棒性;仿真结果表明了其有效性。
Industrial robot control system was influenced inevitably by random noise during its running. A RBF neural network and PD combined controller was proposed based on the Kalman filter, the controller was composed of RBF neural network feed forward, PD feed back and Kalman filter. The control property and dynamic performance of RBF neural network and PD composite control have been greatly improved based on Kalman filter. It can fast track the fixed value without overshoots and oscillations based on less control energy, and the white noise was suppressed, thus its adaptability and robustness can be enhanced. The simulation results show the efficiency of the combined controller.
出处
《计算机测量与控制》
CSCD
北大核心
2009年第8期1551-1553,1573,共4页
Computer Measurement &Control
基金
国家863计划资助项目(2006AA10Z259)
