基于RBF神经网络精密定位预测控制研究

Research on precise alignment predictive control based on the RBF neural network

为了实现高速高精度定位,在RBF神经网络单步预测模型的基础上,建立多步预测控制系统,为当前高速定位控制提供操作指导。在定位过程中通过预测位置与实际位置比较,灵活调整定位速度。实验表明神经网络预测控制是实现高速高精度定位的有效手段。

To realize the high-speed and high-precision positioning,a multi-step predictive control system is established on the basis of the single-step predictive model of the RBF neural network,thus providing the operation guide to the current high-speed positioning.During the alignment,the alignment speed is adjustable flexibly by comparing the predictive position with an actual postion.The experiment result shows that the neural-network-based predictive control is an effective means to perform high-speed and high-precision positioning.